Laser batch processing method, equipment, system and medium

Abstract

The embodiment of the invention discloses a laser batch processing method, device and system and a medium. The method comprises the steps of obtaining a plurality of region-of-interest images through recognition processing of a processing region image, and then matching a region-of-interest image corresponding to a reference processing material in the plurality of region-of-interest images with other region-of-interest images, and filling the to-be-processed materials corresponding to the successfully matched region-of-interest images according to the processing elements to generate a processing image, and generating a processing instruction for performing batch processing on the plurality of to-be-processed materials by using the processing image so as to form the processing elements on each to-be-processed material. According to the technical scheme, the efficiency and accuracy of laser batch processing are improved, and the reliability of laser batch processing is high.

Description

Technical Field

[0001] The present application relates to the field of laser processing technology, and in particular to a laser batch processing method, laser processing equipment, system, and computer-readable medium. Background Art

[0002] It is understandable that laser processing equipment is a device that uses laser beams for processing, and it can use laser beams to cut, punch, engrave, and process various materials such as metals, plastics, wood, glass, textiles, etc. In the related art, when using laser processing equipment to realize laser batch processing of materials to be processed, it is usually necessary to manually place and align the materials to be processed. This process is time-consuming and labor-intensive, which reduces the efficiency of laser batch processing, and manual placement and alignment are also prone to errors, which reduces the accuracy of laser batch processing.

[0003] Therefore, how to improve the reliability of laser batch processing is an urgent problem to be solved. Summary of the invention

[0004] To solve the above technical problems, the embodiments of the present application provide a laser batch processing method, a laser batch processing device, a laser processing equipment, a laser processing system, an electronic device, a computer readable storage medium, and a computer program product. The embodiments of the present application improve the efficiency and accuracy of laser batch processing, and the reliability of laser batch processing is high.

[0005] According to one aspect of an embodiment of the present application, a laser batch processing method is provided, the method comprising: obtaining a processing area image obtained by photographing a processing area where a plurality of materials to be processed are placed; obtaining a plurality of region of interest images based on recognition processing results of the processing area image; determining a reference processing material among the plurality of materials to be processed based on positions of processing elements and the plurality of region of interest images; matching the region of interest image corresponding to the reference processing material in the plurality of region of interest images with the remaining region of interest images; filling the processing elements for the materials to be processed corresponding to the successfully matched region of interest images to generate processing images; and generating processing instructions based on the processing images to batch process the plurality of materials to be processed and form the processing elements.

[0006] In another aspect of an embodiment of the present application, a laser batch processing device is provided, the device comprising: an acquisition module, configured to acquire a processing area image obtained by photographing a processing area where a plurality of materials to be processed are placed; an identification and processing module, configured to obtain a plurality of region of interest images based on the identification and processing results of the processing area image; a determination module, configured to determine a reference processing material among the plurality of materials to be processed based on the positions of the processing elements and the plurality of region of interest images; a matching module, configured to match the region of interest image corresponding to the reference processing material in the plurality of region of interest images with the remaining region of interest images; a filling module, configured to fill the processing elements for the materials to be processed corresponding to the successfully matched region of interest images to generate processing images; and a generation module, configured to generate processing instructions based on the processing images to batch process the plurality of materials to be processed and form the processing elements.

[0007] Another aspect of an embodiment of the present application provides a laser processing device, including: a processing head; a communication component, wherein the communication component is used to receive processing instructions obtained according to the steps of the laser batch processing method as described above; and a controller, wherein the controller controls the movement of the processing head to perform laser processing based on the processing instructions.

[0008] Another aspect of an embodiment of the present application provides a laser processing system, comprising: a laser processing device, the laser processing device comprising a base plate and a processing head, the base plate comprising a processing area for placing materials, the processing head being used to move on the processing area; and a terminal device communicating with the laser processing device, the terminal device being used to execute the laser batch processing method as described above.

[0009] Another aspect of an embodiment of the present application provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs, wherein when the one or more programs are executed by the one or more processors, the electronic device implements the laser batch processing method as described above.

[0010] In another aspect of an embodiment of the present application, a computer-readable storage medium is provided, on which computer-readable instructions are stored. When the computer-readable instructions are executed by a processor of a computer, the computer executes the laser batch processing method as described above.

[0011] Another aspect of an embodiment of the present application provides a computer program product, including a computer program, which implements the laser batch processing method as described above when executed by a processor.

[0012] In the technical solution provided in the embodiments of the present application, a plurality of region of interest images are obtained by identifying and processing the processing area image, and then the region of interest image corresponding to the reference processing material in the plurality of region of interest images is matched with the remaining region of interest images, and then the material to be processed corresponding to the successfully matched region of interest image is filled with processing elements to generate a processing image, and then the processing image is used to generate processing instructions for batch processing of the plurality of materials to be processed, so as to form processing elements on each material to be processed; thereby, laser batch processing is completed simply and accurately.

[0013] In this way, laser batch processing is achieved through automation, and there is no need for manual placement and alignment of materials to be processed, which saves manpower and material resources, improves the efficiency of laser batch processing, and avoids the phenomenon that errors are prone to occur when manually placing and aligning materials to be processed, thereby ensuring the consistency of laser batch processing and thus improving the accuracy of laser batch processing; therefore, the reliability of laser batch processing in the technical solution provided in the embodiments of the present application is high. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is a schematic diagram of an exemplary laser processing system of the present application;

[0015] Figure 2 is a flow chart of a laser batch processing method shown in an exemplary embodiment of the present application;

[0016] Figure 3 is a flow chart of a laser batch processing method shown in another exemplary embodiment of the present application;

[0017] Figure 4 is a schematic diagram of an image of a processing area shown in an exemplary embodiment of the present application;

[0018] Figure 5 is a schematic diagram of target detection shown in an exemplary embodiment of the present application;

[0019] Figure 6 is a schematic diagram of image segmentation shown in an exemplary embodiment of the present application;

[0020] Figure 7 is a schematic diagram of an image of a region of interest corresponding to a reference processing material shown in an exemplary embodiment of the present application;

[0021] Figure 8a to Figure 8b is a schematic diagram of laser batch engraving shown in an exemplary embodiment of the present application;

[0022] Fig. 9 is a block diagram of a laser batch processing device shown in an exemplary embodiment of the present application;

[0023] Fig.10 is a schematic diagram of an exemplary laser processing device of the present application;

[0024] Fig.11 It is a structural diagram of a computer system suitable for implementing an electronic device of an embodiment of the present application. DETAILED DESCRIPTION

[0025] Here, exemplary embodiments will be described in detail, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are only examples of devices and methods consistent with some aspects of the present application as detailed in the attached claims.

[0026] The block diagrams shown in the accompanying drawings are merely functional entities and do not necessarily correspond to physically independent entities. That is, these functional entities may be implemented in software form, or in one or more hardware modules or integrated circuits, or in different networks and / or processor devices and / or microcontroller devices.

[0027] The flowcharts shown in the accompanying drawings are only exemplary and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps can be decomposed, and some operations / steps can be combined or partially combined, so the actual execution order may change according to actual conditions.

[0028] The term "multiple" as used in this application refers to two or more than two. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the related objects are in an "or" relationship.

[0029] The terms "first", "second", "third" and "fourth" in the specification, claims and drawings of this application are used to distinguish different objects, rather than to describe a specific order. The terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0030] In the embodiments of the present application, the term "module" or "unit" refers to a computer program or a part of a computer program with a predetermined function, and works together with other related parts to achieve a predetermined goal, and can be implemented in whole or in part by using software, hardware (such as processing circuits or memories) or a combination thereof. Similarly, a processor (or multiple processors or memories) can be used to implement one or more modules or units. In addition, each module or unit can be part of an overall module or unit that includes the function of the module or unit.

[0031] See also Figure 1 , Figure 1 This is a schematic diagram of an exemplary laser processing system of the present application. With the development of laser processing equipment, laser processing equipment has gradually become intelligent and convenient. Users can conveniently use laser processing equipment to achieve laser batch processing on the materials to be processed by controlling the host computer software installed on the terminal device. Figure 1 As shown, the laser processing system 100 includes a terminal device 101 and a laser processing device 102. The terminal device 101 is communicatively connected to the laser processing system 102. The terminal device 101 is installed with host computer software of the laser processing device 102. The host computer software provides a device control interface. The user can use the laser processing device 102 to realize laser batch processing on the material to be processed by performing relevant trigger operations (such as editing processing patterns, setting processing parameters, etc.) on the device control interface.

[0032] In the process of using laser processing equipment to realize laser batch processing of materials to be processed, it is usually necessary to manually place and align the materials to be processed. This process is time-consuming and labor-intensive, reducing the efficiency of laser batch processing, and manual placement and alignment are also prone to errors, reducing the accuracy of laser batch processing. Therefore, in order to improve the efficiency and accuracy of laser batch processing and ensure the reliability of laser batch processing, this application provides a laser batch processing solution.

[0033] In one embodiment of the present application, the laser batch processing method can be executed by the terminal device 101 (specifically, it can be the upper computer software in the terminal device 101), specifically: the terminal device 101 obtains the processing area image obtained by shooting the processing area where multiple materials to be processed are placed; then, based on the recognition processing result of the processing area image, multiple region of interest images are obtained; then, based on the processing elements and the positions of the multiple region of interest images, the reference processing material among the multiple materials to be processed is determined; then, the region of interest image corresponding to the reference processing material in the multiple region of interest images is matched with the remaining region of interest images; then, the materials to be processed corresponding to the successfully matched region of interest images are filled with processing elements to generate processing images; then, processing instructions are generated based on the processing images to batch process the multiple materials to be processed and form processing elements.

[0034] In one embodiment of the present application, the laser batch processing method can also be executed by the laser processing device 102 alone, or the laser processing device 102 and the terminal device 101 can be executed interactively. In actual applications, it can be flexibly adjusted according to specific application scenarios.

[0035] It should be clear that the terminal device 101 includes but is not limited to smart phones, computers (tablet computers, laptops, desktop computers, etc.), smart wearable devices (bracelets, watches, etc.), etc. The laser processing equipment 102 includes but is not limited to laser engraving machines, laser cutting machines, laser printers, etc.; optionally, the laser processing equipment 102 has a camera 1025 for photographing a processing area where a plurality of materials to be processed are placed, so as to obtain an image of the processing area, wherein the number of cameras 1025 can be one or more.

[0036] It should be noted that the number of terminal devices 101 and laser processing devices 102 is only schematic, and any number of terminal devices 101 and laser processing devices 102 may be provided according to actual needs.

[0037] It is understandable that in the specific implementation of this application, user-related data is involved. When the embodiments of this application are applied to specific products or technologies, user permission or consent is required, and the collection, use and processing of relevant data need to comply with relevant laws, regulations and standards of relevant countries and regions.

[0038] The following is a detailed description of various implementation details of the technical solution of the embodiment of the present application:

[0039] See also Figure 2 , Figure 2 FIG. 1 is a flow chart of a laser batch processing method according to an embodiment of the present application. The laser batch processing method can be executed by the terminal device 101 (specifically, it can be the host computer software in the terminal device 101). Figure 2 As shown, the laser batch processing method at least includes S201 to S206, which are described in detail as follows:

[0040] S201, obtaining an image of a processing area obtained by photographing a processing area where a plurality of materials to be processed are placed.

[0041] In the embodiment of the present application, the material to be processed refers to a material that has not been filled with processing elements and has not been laser processed, and the processing area image refers to an image containing multiple materials to be processed.

[0042] In one embodiment of the present application, the process of obtaining an image of the processing area obtained by photographing the processing area where multiple materials to be processed are placed in S201 may include: generating a shooting instruction; sending the shooting instruction to the laser processing equipment, so that the laser processing equipment controls the camera to photograph the processing area where multiple materials to be processed are placed based on the shooting instruction to obtain the processing area image; and receiving the processing area image sent by the laser processing equipment.

[0043] That is, in the optional embodiment, the terminal device sends the generated shooting instruction to the laser processing device. Accordingly, the laser processing device receives the shooting instruction and uses the shooting instruction to control the camera to shoot the processing area where multiple materials to be processed are placed to obtain the processing area image, and sends the processing area image to the terminal device. Accordingly, the terminal device receives the processing area image. In this way, by implementing the optional embodiment, the processing area image can be easily obtained by shooting with the laser processing device, thereby providing strong support for image recognition processing.

[0044] In one embodiment of the present application, the process of obtaining a processing area image obtained by photographing a processing area where multiple materials to be processed are placed in S201 may include: generating acquisition information for the processing area image, and displaying the acquisition information in an editable interface; receiving a processing area image input in the editable interface, where the processing area image is obtained by a user using a photographing device to photograph the processing area where multiple materials to be processed are placed and uploading it.

[0045] That is, in the optional embodiment, the terminal device displays the generated acquisition information, so that after the user views the acquisition information, the user can use a shooting device (such as a camera of the terminal device or a camera of other devices) to shoot the processing area where multiple materials to be processed are placed to obtain an image of the processing area, and send the image of the processing area to the terminal device. Accordingly, the terminal device receives the image of the processing area. In this way, by implementing the optional embodiment, the image of the processing area can be easily obtained by shooting with an external shooting device, thereby providing strong support for image recognition processing.

[0046] S202, obtaining a plurality of region of interest images based on the recognition processing results of the processing area image.

[0047] In the embodiment of the present application, the terminal device acquires the processing area image, and then can perform recognition processing on the processing area image, thereby obtaining multiple region of interest images using the recognition processing results of the processing area image. Among them, it can be to identify the features of interest in the processing area image, and then obtain the region of interest image. The feature of interest can be the material to be processed, or a partial area on the material to be processed, or a specified pattern area on the material to be processed, or a foreground feature in the processing area image, or an object of a specified shape, or a feature automatically recognized based on the content input by the user, etc.

[0048] In one embodiment of the present application, the process of obtaining multiple images of the region of interest based on the recognition processing results of the processing area image in S202 may include: performing target detection on the processing area image to obtain multiple candidate frames located in the processing area image; performing image segmentation on each candidate frame to obtain the image of the region of interest corresponding to each candidate frame.

[0049] That is, in the optional embodiment, the terminal device first performs target detection on the processing area image to obtain multiple candidate frames located in the processing area image, and then performs image segmentation on each candidate frame to obtain an image of the region of interest corresponding to each candidate frame, thereby obtaining multiple images of the region of interest. Image segmentation refers to segmenting the region of interest, and the region of interest can be determined based on the region of the feature of interest in the processing area image; for example, if the feature of interest is a foreground image, the foreground image is segmented. In this way, by implementing the optional embodiment, the target detection of the processing area image and the image segmentation corresponding to the candidate frame obtained by the target detection can intuitively and accurately identify the object in the processing area image, thereby improving the accuracy of object recognition in the processing area image, thereby improving the accuracy of laser batch processing.

[0050] In one embodiment of the present application, the process of performing target detection on a processing area image to obtain multiple candidate frames located in the processing area image may include: performing feature extraction on the processing area image to obtain semantic features corresponding to the processing area image; performing target detection on the processing area image based on the semantic features to generate multiple candidate frames located in the processing area image.

[0051] That is, in an optional embodiment, the terminal device first extracts features from the processing area image to obtain semantic features corresponding to the processing area image, and then uses the semantic features to perform target detection on the processing area image to generate multiple candidate boxes located in the processing area image, thereby obtaining multiple candidate boxes located in the processing area image.

[0052] Among them, in the optional embodiment, the semantic feature refers to the feature representation of the semantic information in the processing area image.

[0053] In the optional embodiment, multiple candidate boxes located in the processing area image will theoretically select each material to be processed, and one of the candidate boxes will select an object (that is, one candidate box selects one material to be processed). In special cases (such as low clarity of the processing area image, etc.), when there are candidate boxes with incomplete materials to be processed, target detection can be performed again, and when there are empty candidate boxes, they can be ignored; in actual applications, the target detection situation can be flexibly adjusted according to the specific application scenario, as long as each material to be processed is selected by the candidate box.

[0054] In this way, by implementing the optional embodiment, the semantic feature of the processing area image, which is a high-level abstract feature, can be used to accurately detect the target, thereby improving the accuracy of the target detection.

[0055] In one embodiment of the present application, the process of extracting features from a processing area image to obtain semantic features corresponding to the processing area image may include: preprocessing the processing area image, the preprocessing including at least one of image size adjustment, contrast enhancement, noise removal, color space conversion, and input format conversion; extracting features from the preprocessed processing area image to obtain semantic features corresponding to the preprocessed processing area image.

[0056] That is, in an optional embodiment, the terminal device first preprocesses the processing area image to obtain a preprocessed processing area image, and then extracts features from the preprocessed processing area image to obtain semantic features corresponding to the preprocessed processing area image.

[0057] Among them, the preprocessing in the optional embodiment includes but is not limited to at least one of image resizing, contrast enhancement, noise removal, color space conversion, and input format conversion. Among them, image resizing is used to obtain a processing area image of a specified image size, which can be specifically increased or reduced in size; contrast enhancement is used to enhance the difference between different areas in the processing area image to improve the visual quality and information transmission capability of the image, which can be specifically implemented by histogram equalization, contrast stretching, histogram matching and other methods; noise removal is used to eliminate or reduce interference signals or pixels introduced by various factors from the processing area image, which can be specifically implemented by spatial domain filtering, frequency domain filtering, nonlinear filtering and other methods; color space conversion is used to adjust and optimize the color information of the processing area image, and common color spaces include RGB (Red, Green, Blue), HSV (Hue, Saturation, Value), Lab (Luminance, color channels a and b), etc.; input format conversion is used to obtain a processing area image of a specified format, and common formats include lossless compression format, lossy compression format, original image format, etc.

[0058] It is understandable that the optical characteristics of the camera may cause the image of the processing area to be distorted; therefore, in an optional embodiment, the image of the processing area may be firstly subjected to distortion correction so that the corrected image of the processing area can be as close as possible to the state in the real scene, and then preprocessed.

[0059] In this way, by implementing the optional embodiment, the accuracy of semantic feature extraction is improved by preprocessing the processing area image and then extracting the semantic features, thereby improving the accuracy of target detection.

[0060] In one embodiment of the present application, target detection is performed on the processing area image by a model with target detection capability to obtain multiple candidate boxes located in the processing area image. Specifically, feature extraction can be performed on the processing area image by a model with target detection capability to obtain semantic features corresponding to the processing area image.

[0061] Optionally, the process of extracting features from the processing area image by a model having target detection capability to obtain semantic features corresponding to the processing area image may include at least two methods:

[0062] Method 1: Input the processing area image into the first target detection model, and perform forward propagation through the convolution layer and the pooling layer in the first target detection model to obtain mapping features, so as to obtain the semantic features of the processing area image.

[0063] It can be understood that in the first method, the processing area image can be converted into a mapping feature by forward propagation through the convolution layer and the pooling layer in the first target detection model, wherein the mapping feature contains semantic information of different layers. Among them, the number of convolution layers and pooling layers in the optional embodiment shall be based on the number contained in the first target detection model. In practical applications, the number of convolution layers and pooling layers can be flexibly adjusted according to the specific application scenario.

[0064] Method 2: Input the processing area image into the second target detection model, obtain the mapping features of the top layer or the middle layer in the second target detection model, and obtain the semantic features of the processing area image.

[0065] It can be understood that the mapping features of the middle layer or the top layer in the second target detection model in the second method capture the features of different image sizes and abstract levels in the processing area image, such as edges, textures, shapes, etc., which are also the manifestation of semantic information. Among them, the specific layer performance of the middle layer or the top layer in the optional embodiment shall be based on the layers included in the second target detection model. In practical applications, the middle layer or the top layer can be flexibly adjusted according to the specific application scenario.

[0066] Optionally, the process of extracting features of the processing area image by a model with target detection capability to obtain semantic features corresponding to the processing area image may include: preprocessing the processing area image by a model with target detection capability, and extracting features of the preprocessed processing area image to obtain semantic features corresponding to the preprocessed processing area image.

[0067] That is, in the optional embodiment, the terminal device first preprocesses the processing area image through the model with target detection capability, and then extracts features from the preprocessed processing area image to obtain semantic features corresponding to the preprocessed processing area image. For the relevant content of the preprocessing, please refer to the aforementioned embodiment, which will not be repeated here. In this way, by implementing the optional embodiment, the semantic features corresponding to the processing area image can be easily and accurately obtained using the model with target detection capability, thereby improving the efficiency and accuracy of target detection for the processing area image.

[0068] In one embodiment of the present application, target detection is performed on the processing area image by a model with target detection capability to obtain multiple candidate boxes located in the processing area image. Specifically, target detection can be performed on the processing area image based on semantic features by a model with target detection capability to generate multiple candidate boxes located in the processing area image.

[0069] Optionally, the process of performing target detection on the processing area image based on semantic features by a model with target detection capability to generate multiple candidate boxes located in the processing area image may include: inputting the semantic features into a third target detection model, generating multiple candidate boxes in the processing area image using a sliding window or a region proposal network through the third target detection model, and labeling the categories of the multiple candidate boxes.

[0070] It can be understood that the third target detection model can select areas that may contain objects in the processing area image through a sliding window or a region proposal network, that is, generate multiple candidate boxes; at the same time, it can also identify the category of the object selected by each candidate box, where the category can specifically be the material type and size of the object; for example, the categories of multiple candidate boxes are wood, metal, plastic, etc., or the categories of multiple candidate boxes are A wood, B wood, C wood, etc. In actual applications, the coarse and fine granularity of object category recognition can be flexibly adjusted according to the specific application scenario.

[0071] In this way, by implementing the optional embodiment, the third target detection model can be used to easily and accurately generate candidate boxes, thereby improving the efficiency and accuracy of target detection.

[0072] It should be clear that the first target detection model, the second target detection model, and the third target detection model in the aforementioned embodiments can be the same model with target detection capabilities, or different models with target detection capabilities. In practical applications, they can be flexibly adjusted according to specific application scenarios.

[0073] In one embodiment of the present application, the training process of a model with target detection capability may include: obtaining a sample detection image group, the sample detection image group including a first sample detection image and a second sample detection image, the second sample detection image being annotated with multiple candidate boxes relative to the first sample detection image; inputting the first sample detection image into the detection model to be trained to obtain a sample detection result of the first sample detection image; adjusting the parameters of the detection model to be trained by the loss value between the sample detection result and the second sample detection image to obtain a model with target detection capability.

[0074] That is, in the optional embodiment, the terminal device adjusts the parameters of the detection model to be trained through the sample detection image group including the first sample detection image and the second sample detection image, thereby training a model with target detection capability. In this way, by implementing the optional embodiment, a model with target detection capability can be trained simply and accurately, thereby providing strong support for target detection of processing area images.

[0075] In one embodiment of the present application, the process of performing image segmentation on each candidate frame to obtain an image of a region of interest corresponding to each candidate frame may include: intercepting an image corresponding to each candidate frame from the processing area image to obtain multiple candidate frame images; performing image segmentation on each candidate frame image based on the pixels contained in each candidate frame image to obtain an image of a region of interest corresponding to each candidate frame image.

[0076] That is, in the optional embodiment, the terminal device first captures the image corresponding to each candidate frame from the processing area image to obtain multiple candidate frame images, and then uses the pixels contained in each candidate frame image to perform image segmentation on each candidate frame image, thereby obtaining the region of interest image corresponding to each candidate frame image. In this way, by implementing the optional embodiment, the multiple candidate frame images are segmented separately, and the region of interest images corresponding to the multiple candidate frame images can be accurately obtained, thereby improving the accuracy of image segmentation.

[0077] In one embodiment of the present application, image segmentation is performed on each candidate box by a model with image segmentation capabilities to obtain an image of the region of interest corresponding to each candidate box. Specifically, the image corresponding to each candidate box can be intercepted from the processing area image by a model with image segmentation capabilities to obtain multiple candidate box images.

[0078] Optionally, the process of obtaining multiple candidate frame images by intercepting the image corresponding to each candidate frame from the processing area image through a model with image segmentation capabilities may include: inputting the processing area image containing multiple candidate frames into a first target segmentation model, and cropping the image corresponding to each candidate frame from the processing area image based on the pixel points corresponding to the boundary of each candidate frame through the first target segmentation model to obtain multiple candidate frame images.

[0079] It can be understood that the first target segmentation model can intercept the image corresponding to each candidate box through the pixel points corresponding to the boundary of each candidate box, thereby obtaining multiple candidate box images.

[0080] In this way, by implementing the optional embodiment, the candidate frame image can be easily and accurately obtained using the first target segmentation model, thereby improving the efficiency and accuracy of image segmentation.

[0081] In one embodiment of the present application, image segmentation is performed on each candidate frame by a model with image segmentation capability to obtain an image of a region of interest corresponding to each candidate frame. Specifically, each candidate frame image can be segmented based on the pixels contained in each candidate frame image by a model with image segmentation capability to obtain an image of a region of interest corresponding to each candidate frame image.

[0082] Optionally, the process of segmenting each candidate frame image based on the pixels contained in each candidate frame image through a model with image segmentation capabilities to obtain the image of the region of interest corresponding to each candidate frame image may include: inputting multiple candidate frame images into a second target segmentation model, binarizing the pixels contained in each candidate frame image through the second target segmentation model to obtain a mask image corresponding to each candidate frame image, and extracting the image of the region of interest corresponding to each candidate frame image from each candidate frame image based on the mask image corresponding to each candidate frame image.

[0083] It can be understood that the second target segmentation model can binarize the pixels contained in each candidate frame image to obtain a mask image corresponding to each candidate frame image, and use the mask image corresponding to each candidate frame image to extract each candidate frame image, thereby obtaining an area of ​​interest image corresponding to each candidate frame image.

[0084] In this way, by implementing the optional embodiment, the second target segmentation model can be used to easily and accurately obtain the region of interest image corresponding to the candidate frame image, thereby improving the efficiency and accuracy of image segmentation.

[0085] It should be clear that the first target segmentation model and the second target segmentation model in the aforementioned embodiment can be the same model with image segmentation capabilities, or different models with image segmentation capabilities. In practical applications, they can be flexibly adjusted according to specific application scenarios.

[0086] In one embodiment of the present application, the training process of a model with image segmentation capability may include: obtaining a sample segmentation image group, the sample segmentation image group including a first sample segmentation image and a second sample segmentation image, the second sample segmentation image being annotated with a foreground area relative to the first sample segmentation image; inputting the first sample segmentation image into the segmentation model to be trained to obtain a sample segmentation result of the first sample segmentation image; adjusting the parameters of the segmentation model to be trained by the loss value between the sample segmentation result and the second sample segmentation image to obtain a model with image segmentation capability.

[0087] That is, in the optional embodiment, the terminal device adjusts the parameters of the segmentation model to be trained through the sample segmentation image group including the first sample segmentation image and the second sample segmentation image, thereby training a model with image segmentation capability. In this way, by implementing the optional embodiment, a model with image segmentation capability can be trained simply and accurately, thereby providing strong support for the segmentation of the candidate frame image.

[0088] S203, determining a reference processing material among a plurality of materials to be processed based on the processing elements and the positions of the plurality of images of the regions of interest.

[0089] In the embodiment of the present application, the terminal device obtains multiple region of interest images, and then can use the processing elements and the positions of the multiple region of interest images to determine the reference processing material among the multiple materials to be processed.

[0090] In the embodiment of the present application, the reference processing material refers to a processing material that has been filled with elements (such as patterns) among multiple materials to be processed, and is used as a reference for filling processing elements in other materials to be processed among the multiple materials to be processed during laser batch processing. The materials may or may not have been laser processed. Usually, the other materials to be processed are the same or similar in material type, size, etc. to the reference processing material.

[0091] In one embodiment of the present application, the process of determining a reference processing material among multiple materials to be processed based on the positions of the processing elements and multiple region of interest images in S203 may include: matching the position of the processing element in the processing area image with the position of each region of interest image in the processing area image; and using the material to be processed corresponding to the successfully matched region of interest image as the reference processing material.

[0092] That is, in an optional embodiment, the terminal device matches the position of the processing element in the processing area image with the position of each area of ​​interest image in the processing area image, and uses the to-be-processed material corresponding to the successfully matched area of ​​interest image as the reference processing material.

[0093] Among them, the specific form of the position in the optional embodiment is coordinates. Optionally, the matching between positions can be to calculate the distance between coordinates, and specifically, the minimum distance is determined as a successful match. Optionally, the position of the processing element in the processing area image can be the coordinates of the center of the processing element in the processing area image, and the position of the region of interest image in the processing area image can be the coordinate interval of the area surrounded by the outer contour of the region of interest image. If the coordinates corresponding to the processing element are located in the coordinate interval of the area surrounded by the outer contour of the region of interest image, it is determined to be a successful match.

[0094] In this way, by implementing the optional embodiment, a reference processing material can be easily and accurately determined from a plurality of materials to be processed through position matching.

[0095] In one embodiment of the present application, before the process of matching the position of the processing element in the processing area image with the position of each area of ​​interest image in the processing area image, it can also include: displaying the processing area image; in response to a filling operation for the material to be processed specified in the processing area image, filling the processing element on the material to be processed specified in the processing image to obtain a processing area image containing the processing element.

[0096] That is, in the optional embodiment, the terminal device displays the acquired processing area image, so that after the user views the processing area image, a filling operation can be issued for the material to be processed specified in the processing area image. Accordingly, the terminal device receives the filling operation, and in response to the filling operation, fills the processing element on the material to be processed specified in the processing image, thereby obtaining a processing area image containing the processing element, so that the position of the processing element in the processing area image can be determined, and then the position of the processing element in the processing area image is matched with the position of each region of interest image in the processing area image. In this way, by implementing the optional embodiment, the processing area image containing the processing element can be easily and accurately obtained by using the interaction with the user.

[0097] S204, matching the region of interest image corresponding to the reference processing material among the plurality of region of interest images with the remaining region of interest images.

[0098] In the embodiment of the present application, the terminal device determines a reference processing material among multiple materials to be processed, and then can match the region of interest image corresponding to the reference processing material among the multiple region of interest images with the remaining region of interest images.

[0099] In one embodiment of the present application, the process of matching the region of interest image corresponding to the reference processing material in multiple region of interest images with the remaining region of interest images in S204 may at least include the following methods: matching the region of interest image corresponding to the reference processing material in multiple region of interest images with the remaining region of interest images for object shape; and / or, matching the region of interest image corresponding to the reference processing material in multiple region of interest images with the remaining region of interest images for object outer contour; and / or, matching the region of interest image corresponding to the reference processing material in multiple region of interest images with the remaining region of interest images for object similarity; and / or, matching the region of interest image corresponding to the reference processing material in multiple region of interest images with the remaining region of interest images for object type.

[0100] Among them, in the optional embodiments, object shape matching mainly focuses on the overall shape of the object, such as geometric properties and structural features, and can be specifically implemented by shape descriptor comparison, etc.; object contour matching mainly focuses on the shape of the object contour, and can be specifically implemented by edge detection, etc.; object similarity matching mainly focuses on the degree of similarity in the feature space, and can be specifically implemented by Euclidean distance, Hamming distance, Mahalanobis distance, etc.; object type matching mainly focuses on the object category, and can be specifically implemented by category template detection, etc.

[0101] In this way, by implementing the optional embodiments, there are multiple optional matching methods, the matching flexibility is high, and when more matching methods are combined, the matching accuracy can be greatly improved, which is suitable for scenarios of laser batch processing with various precisions.

[0102] S205, filling the to-be-processed material corresponding to the successfully matched region-of-interest image with processing elements to generate a processed image.

[0103] In the embodiment of the present application, the terminal device matches the region of interest image corresponding to the reference processing material in multiple region of interest images with the remaining region of interest images, and then fills the processing elements of the material to be processed corresponding to the successfully matched region of interest image, thereby generating a processed image.

[0104] For example, PB is used to represent the region of interest image corresponding to each candidate box, where PB = {pb1, pb2, pb3, pb4, pb5}; suppose pb2 is the region of interest image corresponding to the reference processing material, pb1, pb3, pb4, pb5 are the region of interest images corresponding to the materials to be processed, then pb2 is matched with pb1, pb2 is matched with pb3, pb2 is matched with pb4, and pb2 is matched with pb5; suppose the above four matches are all successful, then the materials to be processed corresponding to pb1, pb3, pb4, pb5 are filled with processing elements, thereby generating a processed image.

[0105] In one embodiment of the present application, the process of filling the processing elements of the material to be processed corresponding to the successfully matched image of the region of interest in S205 to generate a processed image may include: taking the material to be processed corresponding to the successfully matched image of the region of interest as the target material to be processed; based on the relative position relationship between the reference processing material and the target material to be processed and the relative position relationship between the reference processing material and the processing elements, filling the target material to be processed with respect to the processing elements to generate a processed image; wherein the relative position relationship between the target material to be processed and the processing elements after filling is the same as the relative position relationship between the reference processing material and the processing elements.

[0106] That is, in an optional embodiment, the terminal device calls the material to be processed corresponding to the successfully matched image of the region of interest the target material to be processed. Usually, there are multiple target materials to be processed. Then, the target material to be processed is filled with processing elements by using the relative position relationship between the reference processing material and the target material to be processed, and the relative position relationship between the reference processing material and the processing elements, thereby generating a processed image.

[0107] As described in the above embodiments, a plurality of matching methods may be combined; therefore, when a plurality of matching methods are combined, the image matching of the region of interest is determined to be successful only when each matching method indicates a successful match.

[0108] For example, the image of the region of interest corresponding to the reference processing material is matched with the other images of the region of interest for object shape matching, object outer contour matching, object similarity matching, and object type matching. Among them, only when these four matches are all characterized by successful matching, the material to be processed corresponding to the image of the region of interest is correct, and the material to be processed corresponding to the image of the region of interest can be used as the target material to be processed; and when at least one of the four matches fails to represent the matching, the material to be processed corresponding to the image of the region of interest is wrong, and the material to be processed corresponding to the image of the region of interest cannot be used as the target material to be processed.

[0109] In this way, by implementing the optional embodiment, a processing image can be generated simply and accurately by utilizing the relative position relationship between the reference processing material and the target material to be processed, and the relative position relationship between the reference processing material and the processing element.

[0110] In one embodiment of the present application, the relative position relationship between the reference processing material and the target processing material may include: the position of the target processing material in the processing area image, and the rotation angle of the target processing material relative to the reference processing material. For example, assuming that the target processing material is a 20 cm × 20 cm square wood A, and the reference processing material is a 20 cm × 20 cm square wood B, at this time, the position of the square wood A in the processing area image includes vertex coordinates (such as coordinates of four vertices), and the rotation angle of the square wood A relative to the square wood B is 90 degrees counterclockwise.

[0111] In one embodiment of the present application, the relative positional relationship between the reference processing material and the processing element may include: the position and rotation angle of the processing element on the reference processing material. For example, assuming that the target material to be processed is a square wood A of 20 cm×20 cm, and the processing element is an equilateral triangle pattern with a side length of 10 cm, the position of the equilateral triangle pattern in the square wood A includes vertex coordinates (e.g., coordinates of three vertices), and the rotation angle of the equilateral triangle pattern relative to the square wood A is that a vertex of the triangle coincides with the upper left corner vertex of the square wood A (e.g., rotated 90 degrees counterclockwise).

[0112] Accordingly, the process of filling the target material to be processed with processing elements based on the relative position relationship between the reference processing material and the target material to be processed, and the relative position relationship between the reference processing material and the processing elements to generate a processing image may include: determining the position and rotation angle of the processing elements on the target material to be processed based on the position of the target material to be processed in the processing area image, the rotation angle of the target material to be processed relative to the reference processing material, and the position and rotation angle of the processing elements on the reference processing material; and filling the target material to be processed with processing elements based on the position and rotation angle of the processing elements on the target material to be processed to generate a processing image.

[0113] That is, in the optional embodiment, the terminal device first uses the position of the target material to be processed in the processing area image, the rotation angle of the target material to be processed relative to the reference material to be processed, and the position and rotation angle of the processing element on the reference material to determine the position and rotation angle of the processing element on the target material to be processed, and then uses the position and rotation angle of the processing element on the target material to be processed to fill the target material to be processed with the processing element, thereby generating a processed image. In this way, by implementing the optional embodiment, the position and rotation angle of the processing element on the target material to be processed can be easily and accurately determined by using the position of the target material to be processed in the processing area image, the rotation angle of the target material to be processed relative to the reference material to be processed, and the position and rotation angle of the processing element on the reference material to provide strong support for the generation of the processed image.

[0114] In one embodiment of the present application, after the process of generating the processed image, the process may further include: displaying the processed image to provide a preview of the laser processing effect.

[0115] S206, generating a processing instruction based on the processing image, so as to batch process a plurality of materials to be processed and form processing elements.

[0116] In the embodiment of the present application, the terminal device obtains a processing image, and then can use the processing image to generate processing instructions to batch process multiple materials to be processed and form processing elements, thereby completing laser batch processing.

[0117] As introduced in the above-mentioned embodiment, the target material to be processed may include multiple materials; therefore, in one embodiment of the present application, the process of generating processing instructions based on the processing image in S206 to batch process multiple materials to be processed and form processing elements may include: generating processing instructions based on the processing image; sending the processing instructions to the laser processing equipment, so that the laser processing equipment performs laser processing on the multiple target materials to be processed based on the processing instructions.

[0118] That is, in the optional embodiment, the terminal device sends the processing instruction generated by the processing image to the laser processing device. Accordingly, the laser processing device receives the processing instruction and uses the processing instruction to control the processing head to perform laser processing on multiple target materials to be processed, so that a processing element is formed on each target material to be processed after processing. In this way, by implementing the optional embodiment, laser batch processing can be easily and accurately realized by controlling the laser processing device to perform processing using the processing instruction.

[0119] In the embodiment of the present application, laser batch processing is achieved through automation, and there is no need for manual placement and alignment of the materials to be processed, which saves manpower and material resources, improves the efficiency of laser batch processing, and also avoids the phenomenon that errors are prone to occur in manual placement and alignment of the materials to be processed, thereby ensuring the consistency of laser batch processing and improving the accuracy of laser batch processing.

[0120] The specific scenarios of the embodiments of the present application are described in detail below:

[0121] See also Figure 3 , Figure 3 FIG. 1 is a flow chart of a laser batch processing method according to an embodiment of the present application. The laser batch processing method can be executed by the terminal device 101 (specifically, it can be the host computer software in the terminal device 101). Figure 3 As shown, the laser batch processing method at least includes S301 to S307, which are described in detail as follows:

[0122] S301, obtaining a processing area image obtained by photographing a processing area where a plurality of materials to be processed are placed, and in response to a filling operation for the materials to be processed specified in the processing area image, filling processing elements on the materials to be processed specified in the processing image to obtain a processing area image containing the processing elements.

[0123] Optionally, the terminal device sends the generated shooting instruction to the laser processing equipment. Accordingly, the laser processing equipment receives the shooting instruction and uses the shooting instruction to control the camera to shoot the processing area where multiple materials to be processed are placed to obtain an image of the processing area, and sends the shot image to the terminal device. Accordingly, the terminal device receives the image of the processing area.

[0124] It can be understood that the processing area image can also be called a frame image.

[0125] See also Figure 4 , is a schematic diagram of the processing area image. Figure 4 As shown, the processing area image includes at least four materials to be processed, and each of the materials to be processed is in a square shape.

[0126] S302, inputting the processing area image into a model with target detection capability to perform target detection, and obtaining a processing area image containing multiple candidate frames.

[0127] Optionally, the model with target detection capability preprocesses the processing area image, and performs forward propagation through convolution layers and pooling layers to obtain mapping features to obtain semantic features of the processing area image, or obtains mapping features of the top layer or middle layer in the model with target detection capability to obtain semantic features of the processing area image, and then uses a sliding window or a region proposal network to generate multiple candidate boxes in the processing area image, and labels the categories of the multiple candidate boxes, and then outputs a candidate box list, wherein the candidate box list includes location information and category information of each candidate box.

[0128] See also Figure 5 , is a schematic diagram of target detection. Figure 5 As shown, target detection is performed on the processing area image by a model with target detection capability, and multiple candidate boxes are generated in the processing area image.

[0129] S303, inputting the processing area image containing multiple candidate frames into a model with image segmentation capability to perform image segmentation, and obtaining an image of a region of interest corresponding to each candidate frame.

[0130] Optionally, the aforementioned candidate box list and processing area image are input into a model with image segmentation capability, and the image corresponding to each candidate box is cropped from the processing area image using the pixel points corresponding to the boundary of each candidate box by the model with image segmentation capability to obtain multiple candidate box images, and then the pixel points contained in each candidate box image are binarized to obtain a mask image corresponding to each candidate box image, and based on the mask image corresponding to each candidate box image, the foreground area corresponding to each candidate box image is extracted from each candidate box image to obtain an area of ​​interest image corresponding to each candidate box.

[0131] See also Figure 6 , is a schematic diagram of image segmentation. Figure 6 As shown, the processing area image containing multiple candidate frames is segmented by a model with image segmentation capability to obtain an image of a region of interest corresponding to each candidate frame, in which the white area is the foreground area.

[0132] It should be noted that the model with target detection capability and the model with image segmentation capability can be integrated in the same model or can be independent models. In practical applications, they can be flexibly adjusted according to specific application scenarios.

[0133] S304, obtaining the position of the processing element in the processing area image, and matching the position of the processing element in the processing area image with the position of each region of interest image in the processing area image, and taking the to-be-processed material corresponding to the successfully matched region of interest image as the reference processing material.

[0134] As described in the above-mentioned embodiment, once the processing area image containing the processing element is obtained, the position of the processing element in the processing area image can be obtained.

[0135] As described in the above embodiment, through position matching, the material to be processed in the upper left corner of the four materials to be processed is the reference material to be processed. Accordingly, the image of the region of interest corresponding to the reference material to be processed is as follows: Figure 7 shown.

[0136] S305, matching the region of interest image corresponding to the reference processing material with the region of interest image corresponding to each of the remaining materials to be processed.

[0137] Optionally, the region of interest image corresponding to the reference processing material is matched with the region of interest images corresponding to each of the remaining materials to be processed in terms of object shape matching, object outer contour matching, object similarity matching, and object type matching.

[0138] Optionally, the region of interest image corresponding to the reference processing material is matched with the region of interest image corresponding to each of the remaining materials to be processed respectively through a model with matching capability.

[0139] S306, filling the to-be-processed material corresponding to the successfully matched region-of-interest image with processing elements to generate a processed image.

[0140] Optionally, the material to be processed corresponding to the successfully matched image of the region of interest is used as the target material to be processed, and there are multiple target processing materials. Based on the relative position relationship between the reference processing material and the target material, and the relative position relationship between the reference processing material and the processing element, each target material to be processed is filled with processing elements to generate a processed image.

[0141] For easier understanding, see Figure 8a and Figure 8b , where the processing area includes multiple objects, the goal of this laser batch processing is to engrave a triangular pattern on an oval-shaped piece of wood. Figure 8a As shown in the figure, there are four oval-shaped woods. The user first fills a triangular pattern on a certain oval-shaped wood as a reference processing material, and then fills triangular patterns on the other three oval-shaped woods to obtain the following Figure 8b The processed image shown.

[0142] S307, laser processing is performed on the material to be processed filled with elements based on the processing image control to complete laser batch processing.

[0143] Optionally, the terminal device generates processing instructions using the processing image and sends the processing instructions to the laser processing equipment. Accordingly, the laser processing equipment receives the processing instructions and uses the processing instructions to control the processing head to perform laser processing on multiple target materials to be processed, so that a processing element is formed on each target material to be processed after processing.

[0144] Following the above, we can get Figure 8b As shown in the example of the processing image, it is necessary to control the laser processing equipment to use the processing image to respectively engrave on four elliptical shaped woods filled with triangular patterns to obtain four elliptical shaped woods engraved with triangular patterns, thereby completing laser batch processing.

[0145] In the embodiments of the present application, corresponding processing in the laser batch processing process is realized by using corresponding models in artificial intelligence technology (AI), so that the batch processing function of the laser processing equipment has a higher degree of intelligence and automation, and is suitable for many laser batch processing scenarios.

[0146] Fig. 9 : is a block diagram of a laser batch processing device shown in an exemplary embodiment of the present application. The device includes:

[0147] An acquisition module 901 is configured to acquire an image of a processing area obtained by photographing a processing area where a plurality of materials to be processed are placed;

[0148] The recognition processing module 902 is configured to obtain a plurality of region of interest images based on the recognition processing results of the processing area image;

[0149] A determination module 903 is configured to determine a reference processing material among a plurality of materials to be processed based on the positions of the processing elements and the plurality of images of the regions of interest;

[0150] A matching module 904 is configured to match the region of interest image corresponding to the reference processed material among the plurality of region of interest images with the remaining region of interest images;

[0151] A filling module 905 is configured to fill the to-be-processed material corresponding to the successfully matched region-of-interest image with respect to processing elements to generate a processed image;

[0152] The generating module 906 is configured to generate a processing instruction based on the processing image, so as to perform batch processing on a plurality of materials to be processed and form processing elements.

[0153] In one embodiment of the present application, based on the aforementioned scheme, the recognition and processing module 902 is specifically configured as follows: performing target detection on the processing area image to obtain multiple candidate frames located in the processing area image; performing image segmentation on each candidate frame to obtain an image of the area of ​​interest corresponding to each candidate frame.

[0154] In one embodiment of the present application, based on the aforementioned scheme, the recognition processing module 902 is further specifically configured to: perform feature extraction on the processing area image to obtain semantic features corresponding to the processing area image; perform target detection on the processing area image based on the semantic features to generate multiple candidate boxes located in the processing area image.

[0155] In one embodiment of the present application, based on the aforementioned scheme, the recognition processing module 902 is further specifically configured to: preprocess the processing area image, the preprocessing including at least one of image size adjustment, contrast enhancement, noise removal, color space conversion, and input format conversion; perform feature extraction on the preprocessed processing area image to obtain semantic features corresponding to the preprocessed processing area image.

[0156] In one embodiment of the present application, based on the aforementioned scheme, the recognition processing module 902 is further specifically configured to: input the processing area image into the first target detection model, perform forward propagation through the convolution layer and the pooling layer in the first target detection model to obtain mapping features, and use the mapping features as semantic features of the processing area image; or input the processing area image into the second target detection model, obtain the mapping features of the top layer or the middle layer in the second target detection model, and use the mapping features as semantic features of the processing area image.

[0157] In one embodiment of the present application, based on the aforementioned scheme, the recognition processing module 902 is further specifically configured to: input the semantic features into the third target detection model, generate multiple candidate boxes on the processing area image through the third target detection model using a sliding window or a region proposal network, and mark the categories of the multiple candidate boxes.

[0158] In one embodiment of the present application, based on the aforementioned scheme, the recognition and processing module 902 is further specifically configured as follows: capturing the image corresponding to each candidate frame from the processing area image to obtain multiple candidate frame images; segmenting each candidate frame image based on the pixels contained in each candidate frame image to obtain an image of the area of ​​interest corresponding to each candidate frame image.

[0159] In one embodiment of the present application, based on the aforementioned scheme, the recognition and processing module 902 is further specifically configured to: input the processing area image containing multiple candidate boxes into the first target segmentation model, and through the first target segmentation model, based on the pixel points corresponding to the boundary of each candidate box, crop the image corresponding to each candidate box from the processing area image to obtain multiple candidate box images.

[0160] In one embodiment of the present application, based on the aforementioned scheme, the recognition processing module 902 is further specifically configured to: input multiple candidate frame images into the second target segmentation model, binarize the pixels contained in each candidate frame image through the second target segmentation model to obtain a mask image corresponding to each candidate frame image, and based on the mask image corresponding to each candidate frame image, extract the region of interest image corresponding to each candidate frame image from each candidate frame image.

[0161] In one embodiment of the present application, based on the aforementioned scheme, the determination module 903 is specifically configured as follows: matching the position of the processing element in the processing area image with the position of each image of the region of interest in the processing area image; and using the material to be processed corresponding to the successfully matched image of the region of interest as the reference processing material.

[0162] In one embodiment of the present application, based on the aforementioned scheme, the determination module 903 is further specifically configured to: display a processing area image in an editable interface; in response to a filling operation for the material to be processed specified in the processing area image, fill in processing elements on the material to be processed specified in the processing image to obtain a processing area image containing processing elements.

[0163] In one embodiment of the present application, based on the aforementioned scheme, the matching module 904 is specifically configured as follows: performing object shape matching on the region of interest image corresponding to the reference processing material in the multiple region of interest images with the remaining region of interest images; and / or performing object outer contour matching on the region of interest image corresponding to the reference processing material in the multiple region of interest images with the remaining region of interest images; and / or performing object similarity matching on the region of interest image corresponding to the reference processing material in the multiple region of interest images with the remaining region of interest images; and / or performing object type matching on the region of interest image corresponding to the reference processing material in the multiple region of interest images with the remaining region of interest images.

[0164] In one embodiment of the present application, based on the aforementioned scheme, the filling module 905 is specifically configured as follows: the material to be processed corresponding to the successfully matched image of the region of interest is used as the target material to be processed; based on the relative position relationship between the reference processing material and the target material to be processed, and the relative position relationship between the reference processing material and the processing element, the target material to be processed is filled with respect to the processing element to generate a processed image; wherein the relative position relationship between the target material to be processed and the processing element after filling is the same as the relative position relationship between the reference processing material and the processing element.

[0165] In one embodiment of the present application, based on the above scheme, the relative position relationship between the reference processing material and the target material to be processed includes: the position of the target material to be processed in the processing area image, and the rotation angle of the target material to be processed relative to the reference processing material; the relative position relationship between the reference processing material and the processing element includes: the position and rotation angle of the processing element on the reference processing material. The filling module 905 is specifically configured to: determine the position and rotation angle of the processing element on the target material to be processed based on the position of the target material to be processed in the processing area image, the rotation angle of the target material to be processed relative to the reference processing material, and the position and rotation angle of the processing element on the reference processing material; fill the target material to be processed with respect to the processing element based on the position and rotation angle of the processing element on the target material to be processed, and generate a processing image.

[0166] It should be noted that the laser batch processing device provided in the above embodiment and the laser batch processing method provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs operations has been described in detail in the method embodiment and will not be repeated here.

[0167] like Fig.10 As shown, the embodiment of the present application further provides a laser processing device 102, including a processing head 1021, a communication component 1022 and a controller 1023. The processing head 1021 is used to output laser to process the material to be processed, the communication component 1022 is used to receive the processing instructions obtained by step identification of the laser batch processing method provided in each of the above embodiments; the controller 1023 is used to control the movement of the processing head based on the processing instructions to perform laser processing.

[0168] Please refer to Figure 1 The embodiment of the present application also provides a laser processing system 100, including a laser processing device 102 and a terminal device 101. The terminal device 101 is communicatively connected to the laser processing device 102. The terminal device 101 can implement the laser batch processing method provided in the above-mentioned embodiments to generate processing instructions and send them to the laser processing device 102. The laser processing device 102 can receive the processing instructions and execute them to realize batch processing of multiple materials to be processed.

[0169] The laser processing device 102 may include a base plate 1024 , and the base plate 1024 includes a processing area for placing the material to be processed. The laser output by the processing head 1021 may move on the processing area to process the material to be processed.

[0170] In some embodiments, the laser processing device 102 may further include a camera 1025 for collecting images of multiple materials to be processed in the processing area, and the laser processing device 101 may upload the images collected by the camera 1025 to the terminal device 101 via the communication component 1022. The controller 1023 may also control the camera 1025 to collect images based on the processing instructions.

[0171] The laser processing equipment 102 may also include other related functional components, which will not be described in detail here.

[0172] An embodiment of the present application also provides an electronic device, comprising: one or more processors; a memory for storing one or more programs, and when the one or more programs are executed by one or more processors, the electronic device implements the laser batch processing method provided in the above-mentioned embodiments.

[0173] Fig.11 It is a schematic diagram of the structure of a computer system suitable for implementing an electronic device of an embodiment of the present application. It should be noted that: Fig.11 The computer system 1000 of the electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present application.

[0174] like Fig.11 As shown, the computer system 1000 includes a central processing unit (CPU) 1001, which can perform various appropriate actions and processes according to the program stored in the read-only memory (ROM) 1002 or the program loaded from the storage part 1008 to the random access memory (RAM) 1003, such as executing the method in the above embodiment. In the RAM 1003, various programs and data required for system operation are also stored. The CPU 1001, the ROM 1002 and the RAM 1003 are connected to each other through the bus 1004. The input / output (I / O) interface 1005 is also connected to the bus 1004.

[0175] The following components are connected to the I / O interface 1005: an input section 1006 including a keyboard, a mouse, etc.; an output section 1007 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 1008 including a hard disk, etc.; and a communication section 1009 including a network interface card such as a local area network (LAN) card, a modem, etc. The communication section 1009 performs communication processing via a network such as the Internet. A drive 1010 is also connected to the I / O interface 1005 as needed. A removable medium 1011, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 1010 as needed, so that a computer program read therefrom is installed into the storage section 1008 as needed.

[0176] In particular, according to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication section 1009, and / or installed from a removable medium 1011. When the computer program is executed by a central processing unit (CPU) 1001, various functions defined in the system of the present application are executed.

[0177] It should be noted that the computer-readable medium shown in the embodiment of the present application can be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. The computer program contained in the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wireless, wired, etc., or any suitable combination of the above.

[0178] The flowchart and block diagram in the accompanying drawings illustrate the possible architecture, functions and operations of the system, method and computer program product according to various embodiments of the present application. Wherein, each box in the flowchart or block diagram can represent a module, a program segment, or a part of the code, and the above-mentioned module, program segment, or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0179] The units involved in the embodiments described in this application may be implemented by software or hardware, and the units described may also be set in a processor. The names of these units do not, in some cases, constitute limitations on the units themselves.

[0180] Another aspect of the present application further provides a computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the laser batch processing method as described above is implemented. The computer-readable storage medium may be included in the electronic device described in the above embodiment, or may exist independently without being assembled into the electronic device.

[0181] Another aspect of the present application also provides a computer program product or a computer program, which includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium, and the processor executes the computer instructions, so that the computer device executes the laser batch processing method provided in each of the above embodiments.

[0182] The above content is only a preferred exemplary embodiment of the present application and is not intended to limit the implementation scheme of the present application. A person skilled in the art can easily make corresponding changes or modifications based on the main concept and spirit of the present application. Therefore, the scope of protection of the present application shall be based on the scope of protection required by the claims.

Claims

1. A laser batch processing method, characterized in that: include: Acquire a processing area image obtained by photographing a processing area where a plurality of materials to be processed are placed; Based on the recognition processing results of the processing area image, a plurality of region of interest images are obtained; Determining a reference processing material among the plurality of materials to be processed based on the processing elements and the positions of the plurality of images of the regions of interest; Matching the region of interest image corresponding to the reference processed material among the plurality of region of interest images with the remaining region of interest images; Filling the material to be processed corresponding to the successfully matched region of interest image with respect to the processing elements to generate a processed image; A processing instruction is generated based on the processing image to batch process the plurality of materials to be processed and form the processing elements.

2. The method according to claim 1, characterized in that The recognition processing result based on the processing area image obtains a plurality of region of interest images, including: Performing target detection on the processing area image to obtain a plurality of candidate frames located in the processing area image; Perform image segmentation on each candidate frame to obtain an image of a region of interest corresponding to each candidate frame.

3. The method according to claim 2, characterized in that The performing target detection on the processing area image to obtain a plurality of candidate frames located in the processing area image includes: Preprocessing the processing area image, wherein the preprocessing includes at least one of image size adjustment, contrast enhancement, noise removal, color space conversion, and input format conversion; Performing feature extraction on the preprocessed processing area image to obtain semantic features corresponding to the preprocessed processing area image; Based on the semantic features, target detection is performed on the processing area image to generate multiple candidate boxes located in the processing area image.

4. The method according to claim 3, characterized in that The step of extracting features from the preprocessed processing area image to obtain semantic features corresponding to the preprocessed processing area image includes: Inputting the preprocessed processing area image into a first target detection model, performing forward propagation through the convolution layer and the pooling layer in the first target detection model to obtain mapping features, and using the mapping features as semantic features of the processing area image; or The preprocessed processing area image is input into a second target detection model, mapping features of a top layer or a middle layer in the second target detection model are obtained, and the mapping features are used as semantic features of the processing area image.

5. The method according to claim 3, characterized in that: The performing target detection on the processing area image based on the semantic features to generate a plurality of candidate frames located in the processing area image includes: The semantic features are input into a third target detection model, and a plurality of candidate boxes are generated on the processing area image by using a sliding window or a region proposal network through the third target detection model, and the categories of the plurality of candidate boxes are marked.

6. The method according to claim 2, characterized in that The performing image segmentation on each candidate frame to obtain an image of a region of interest corresponding to each candidate frame includes: Intercepting an image corresponding to each candidate frame from the processing area image to obtain a plurality of candidate frame images; Image segmentation is performed on each candidate frame image based on the pixels contained in each candidate frame image to obtain a region of interest image corresponding to each candidate frame image.

7. The method according to claim 6, characterized in that The step of intercepting an image corresponding to each candidate frame from the processing area image to obtain a plurality of candidate frame images includes: Inputting the processing area image containing the multiple candidate frames into the first target segmentation model, and using the first target segmentation model to crop the image corresponding to each candidate frame from the processing area image based on the pixel points corresponding to the boundary of each candidate frame, so as to obtain multiple candidate frame images; and / or, The performing image segmentation on each candidate frame image based on the pixels contained in each candidate frame image to obtain a region of interest image corresponding to each candidate frame image includes: The multiple candidate frame images are input into a second target segmentation model, and the pixel points contained in each candidate frame image are binarized by the second target segmentation model to obtain a mask image corresponding to each candidate frame image, and based on the mask image corresponding to each candidate frame image, an area of ​​interest image corresponding to each candidate frame image is extracted from each candidate frame image.

8. The method according to any one of claims 1 to 7, characterized in that The step of determining a reference processing material among the plurality of materials to be processed based on the processing element and the positions of the plurality of images of the regions of interest comprises: Matching the position of the processing element in the processing area image with the position of each region of interest image in the processing area image; The material to be processed corresponding to the successfully matched region of interest image is used as a reference processing material.

9. The method according to claim 8, characterized in that Before matching the position of the processing element in the processing area image with the position of each region of interest image in the processing area image, the method further includes: Displaying the processing area image; In response to a filling operation on the material to be processed specified in the processing area image, the processing elements are filled on the material to be processed specified in the processing image to obtain the processing area image containing the processing elements.

10. The method according to any one of claims 1 to 7, characterized in that The filling of the processing elements of the material to be processed corresponding to the successfully matched region of interest image to generate a processed image includes: The material to be processed corresponding to the successfully matched region of interest image is used as the target material to be processed; Based on the relative position relationship between the reference processing material and the target material to be processed, and the relative position relationship between the reference processing material and the processing element, the target material to be processed is filled with respect to the processing element to generate a processing image; wherein the relative position relationship between the target material to be processed and the processing element after filling is the same as the relative position relationship between the reference processing material and the processing element.

11. The method according to claim 10, characterized in that The relative position relationship between the reference processing material and the target processing material includes: the position of the target processing material in the processing area image, and the rotation angle of the target processing material relative to the reference processing material; The relative positional relationship between the reference processing material and the processing element includes: the position and rotation angle of the processing element on the reference processing material; The step of filling the target material to be processed with respect to the processing element based on the relative position relationship between the reference processing material and the target material to be processed and the relative position relationship between the reference processing material and the processing element to generate a processing image includes: Determine the position and rotation angle of the processing element on the target material to be processed based on the position of the target material to be processed in the processing area image, the rotation angle of the target material to be processed relative to the reference material to be processed, and the position and rotation angle of the processing element on the reference material to be processed; Based on the position and rotation angle of the processing element on the target material to be processed, the target material to be processed is filled with the processing element to generate a processing image.

12. A laser processing device, characterized in that: include: Processing head; A communication component, the communication component is used to receive a processing instruction obtained according to the steps of the laser batch processing method according to any one of claims 1 to 11; A controller controls the movement of the processing head to perform laser processing based on the processing instruction.

13. A laser processing system, characterized in that: include: A laser processing device, the laser processing device comprising a base plate and a processing head, the base plate comprising a processing area for placing a material, the processing head being used to move over the processing area; and A terminal device that communicates with the laser processing device, wherein the terminal device is used to execute the laser batch processing method according to any one of claims 1 to 11.

14. A computer readable medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the laser batch processing method according to any one of claims 1 to 11 is implemented.

Images