Ultrasonic scanning control method and system based on three-dimensional real-time imaging

Abstract

The invention relates to an ultrasonic scanning control method and system based on three-dimensional real-time imaging, and the method comprises the steps: capturing a three-dimensional image of a current welding region in real time based on image collection equipment, and generating a three-dimensional model corresponding to the current welding region based on a preset image processing algorithm; based on the three-dimensional model, welding seam information corresponding to the current welding area is obtained; based on the weld information, the moving parameters of the ultrasonic scanning probe and the scanning efficiency, the optimal ultrasonic scanning path corresponding to the current welding area is obtained; and based on the optimal ultrasonic scanning path, an ultrasonic scanning probe is controlled to move, and based on ultrasonic signal data collected by the ultrasonic scanning probe in real time, welding seam quality evaluation information corresponding to the current welding area is obtained. According to the method, the welding seam identification precision is improved, the dynamic monitoring of the welding seam quality is realized, the welding seam quality evaluation information can be provided in time, and the real-time evaluation function is beneficial to finding and processing the quality problem in the welding process in time.

Description

Technical Field

[0001] This application relates to the technical field of welding quality inspection, and in particular to an ultrasonic scanning control method and system based on three-dimensional real-time imaging. Background Technology

[0002] With the rapid development of the new energy industry, energy storage liquid cooling plates, as key components, play a crucial role in the thermal management of battery packs. The welding quality of their friction welding directly affects the heat dissipation performance, safety, and service life of the battery pack. Traditional welding quality inspection methods, such as X-ray inspection and penetrant testing, have drawbacks such as low inspection efficiency, high cost, and harm to human health, making it difficult to meet the demands of modern industry for high efficiency, environmental protection, and intelligence. Therefore, there is an urgent need for a highly automated ultrasonic scanning technology solution capable of real-time detection. Summary of the Invention

[0003] To improve the automation level of welding quality inspection in friction welding and achieve real-time ultrasonic scanning inspection, this application provides an ultrasonic scanning control method and system based on three-dimensional real-time imaging.

[0004] In a first aspect, this application provides an ultrasound scanning control method based on three-dimensional real-time imaging, comprising: Based on an image acquisition device, a three-dimensional image of the current welding area is captured in real time, and a three-dimensional model corresponding to the current welding area is generated based on a preset image processing algorithm. The image acquisition device includes a 3D camera and a laser projector. Based on the three-dimensional model, the weld information corresponding to the current welding area is obtained, and the weld information includes the shape, position and size of the weld. Based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency, the optimal ultrasonic scanning path corresponding to the current welding area is obtained. The movement parameters include the physical size and range of motion of the ultrasonic scanning probe, and the scanning efficiency is a parameter that characterizes how fast the ultrasonic scanning task corresponding to the current welding area is completed. Based on the optimal ultrasonic scanning path, the ultrasonic scanning probe is moved, and based on the ultrasonic signal data collected in real time by the ultrasonic scanning probe, the weld quality assessment information corresponding to the current welding area is obtained.

[0005] The beneficial effects of this application are as follows: the three-dimensional model can more realistically reflect the actual situation of the welding area, thereby greatly improving the accuracy of weld identification. By controlling the ultrasonic scanning probe to move along the optimal path and acquiring ultrasonic signal data in real time, dynamic monitoring of weld quality is realized. After processing and analysis, the acquired data can provide weld quality assessment information in real time. The real-time assessment function helps to promptly detect and handle quality problems in the welding process, avoiding safety hazards in subsequent processing and use.

[0006] Furthermore, before controlling the movement of the ultrasound scanning probe based on the optimal ultrasound scanning path, the method further includes: Based on the welding material corresponding to the current welding area, the initial operating parameters of the ultrasonic scanning probe are obtained, including the scanning mode, frequency, and gain of the ultrasonic scanning probe. After controlling the movement of the ultrasound scanning probe based on the optimal ultrasound scanning path, the method further includes: The system acquires environmental information corresponding to the current welding area in real time, and adjusts the initial working parameters based on the environmental information, which includes temperature, humidity, and noise.

[0007] The beneficial effects of adopting the above-mentioned further solutions are as follows: By setting appropriate initial operating parameters of the ultrasonic scanning probe according to the characteristics of the welding material, the targeting and effectiveness of the detection process can be ensured, improving the accuracy and reliability of the detection. By acquiring environmental information in real time and dynamically adjusting the operating parameters of the ultrasonic scanning probe, it is possible to adapt to the detection needs in different environments, improving the adaptability and stability of the detection.

[0008] Furthermore, obtaining the optimal ultrasonic scanning path corresponding to the current welding area based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency includes: Based on a preset path generation algorithm, the weld information, the movement parameters of the ultrasonic scanning probe, and the initial ultrasonic scanning path corresponding to the scanning efficiency are obtained. Based on the weld information, weld quality prediction information for the current welding area is obtained; Based on the weld quality prediction information, the initial ultrasonic scanning path is adjusted to obtain the optimal ultrasonic scanning path.

[0009] The beneficial effects of adopting the above-mentioned further scheme are: by adjusting the ultrasonic scanning path based on weld information and quality prediction information, the detection process can be more accurately targeted at key areas in the weld, helping to reduce the risk of missed and false detections, and improving the accuracy and reliability of the detection. The ultrasonic scanning path can be flexibly adjusted according to the characteristics of different welds and detection requirements, making the ultrasonic scanning control method based on three-dimensional real-time imaging widely applicable and suitable for different types of friction welding quality inspection scenarios for energy storage liquid cooling plates.

[0010] Furthermore, after obtaining the optimal ultrasonic scanning path corresponding to the current welding area based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency, the method further includes: Import the 3D model into the created AR scene, and automatically adjust the position, size, and rotation angle of the 3D model to match the actual current welding area; On the 3D model in the AR scene, annotation information corresponding to the weld information is marked, including color information and texture information; and on the 3D model in the AR scene, a preview line or animation corresponding to the optimal ultrasonic scanning path is drawn. Based on the data synchronization mechanism, the AR scene is updated to the AR display device in real time, so as to provide an interactive interface for users based on the AR display device.

[0011] The beneficial effects of adopting the above-mentioned further solutions are: by utilizing AR technology, information such as 3D models, annotation information, and scan path previews are intuitively displayed to users, enhancing the visualization of the inspection process and helping users to better understand and operate it. By providing an intuitive interactive interface and real-time AR scene updates, the difficulty of operation is reduced, enabling non-professionals to easily perform welding quality inspection.

[0012] Furthermore, the step of obtaining weld quality assessment information corresponding to the current welding area based on the ultrasonic signal data acquired in real time by the ultrasonic scanning probe includes: Based on the key features extracted from the ultrasonic signal data, the weld quality assessment information is obtained; If the weld quality assessment information indicates the presence of weld defects, then obtain the corresponding defect correction suggestions and defect cause analysis information, and generate a defect report based on the defect correction suggestions, defect cause analysis information, defect type, defect location, and defect size.

[0013] The beneficial effects of adopting the above-mentioned further solutions are that defect reports provide strong support for subsequent repairs and improvements. By combining advanced technologies such as deep learning models, an intelligent solution is provided for the friction welding quality inspection of energy storage liquid cooling plates, which helps to promote the intelligent development of related industries.

[0014] Furthermore, the process of obtaining the weld quality assessment information based on the key features extracted from the ultrasonic signal data includes: Based on the welding material and the welding process corresponding to the current welding area, the input features and output categories of the initial deep learning model are adjusted, and a target deep learning model is constructed based on the adjusted input features and output categories. Based on the adjusted input features, the key features are extracted and input into the target deep learning model to obtain the weld quality assessment information output by the target deep learning model.

[0015] The advantages of adopting the above-mentioned further approach are: the input features and output categories of the deep learning model can be flexibly adjusted according to different welding materials and welding processes, thereby adapting to different inspection needs. Through the training and optimization of the deep learning model, key features in ultrasonic signals can be accurately extracted, and weld quality can be accurately evaluated based on these features.

[0016] Furthermore, before adjusting the input features and output category of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area, the method further includes: Determine whether the welding material and welding process corresponding to the initial deep learning model are the same as the welding material and welding process corresponding to the current welding area; If they are different, then the step of adjusting the input features and output category of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area is executed; If they are the same, then the initial deep learning model is used as the target deep learning model corresponding to the current welding area.

[0017] The beneficial effects of adopting the above-mentioned further approach are: by ensuring that the deep learning model matches the welding material and welding process of the current welding area, the accuracy of weld quality assessment can be significantly improved. When the welding material and welding process corresponding to the initial deep learning model are the same as the information of the current welding area, the existing deep learning model can be used directly without additional model adjustment or training, thereby saving time and resources.

[0018] Secondly, this application provides an ultrasonic scanning control system based on three-dimensional real-time imaging, comprising: An image capture module is used to capture a three-dimensional image of the current welding area in real time based on an image acquisition device, and to generate a three-dimensional model corresponding to the current welding area based on a preset image processing algorithm. The image acquisition device includes a 3D camera and a laser projector. The weld acquisition module is used to acquire weld information corresponding to the current welding area based on the three-dimensional model. The weld information includes the shape, position, and size of the weld. The path acquisition module is used to acquire the optimal ultrasonic scanning path corresponding to the current welding area based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency. The movement parameters include the physical size and movement range of the ultrasonic scanning probe, and the scanning efficiency is a parameter that characterizes how fast the ultrasonic scanning task corresponding to the current welding area is completed. The control and movement module is used to control the movement of the ultrasonic scanning probe based on the optimal ultrasonic scanning path, and to obtain weld quality assessment information corresponding to the current welding area based on the ultrasonic signal data collected in real time by the ultrasonic scanning probe.

[0019] Thirdly, this application provides an electronic device, including a processor and a memory, wherein the processor is coupled to the memory; The processor is configured to execute a computer program stored in the memory, causing the electronic device to perform the method as described in any of the first aspects.

[0020] Fourthly, this application provides a computer-readable storage medium including a computer program or instructions that, when executed on a computer, cause the computer to perform the method as described in any of the first aspects. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the ultrasonic scanning control method based on three-dimensional real-time imaging, as described in an embodiment of this application. Figure 2 This is a structural block diagram of an ultrasonic scanning control system based on three-dimensional real-time imaging, according to an embodiment of this application. Figure 3 This is a structural block diagram of an electronic device according to an embodiment of this application. Detailed Implementation

[0022] The present application will be further described in detail below with reference to the accompanying drawings.

[0023] This application provides an ultrasound scanning control method based on three-dimensional real-time imaging. This method can be executed by a device, which can be a server or a terminal device. The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services. The terminal device can be a smartphone, tablet computer, desktop computer, etc., but is not limited to these.

[0024] like Figure 1As shown, an ultrasound scanning control method based on three-dimensional real-time imaging, with electronic equipment as the execution subject, is described in its main process flow as follows (steps S101 to S104): Step S101: Based on the image acquisition device, capture the three-dimensional image of the current welding area in real time, and generate the three-dimensional model corresponding to the current welding area based on the preset image processing algorithm. The image acquisition device includes a 3D camera and a laser projector.

[0025] In this embodiment, by setting up an image acquisition device, three-dimensional images of the welding area can be captured in real time, ensuring that dynamic changes during the welding process are captured. Using a high-precision 3D camera, it is possible to capture the fine three-dimensional structure of the welding area. A laser projector is used to provide additional depth information, helping the 3D camera to more accurately reconstruct the three-dimensional model of the welding area. The 3D camera and laser projector can be fixed in appropriate positions to ensure they can cover the entire welding area.

[0026] In this embodiment, advanced image processing algorithms such as structured light and stereo vision can be used to process the captured three-dimensional image to generate a three-dimensional model of the welding area. This three-dimensional model can include detailed information such as the shape and surface texture of the welding area.

[0027] High-precision 3D cameras and laser projectors capture real-time 3D images of the welding area and generate corresponding 3D models. These 3D models more realistically reflect the actual conditions of the welding area, including the shape, location, and minute details of the weld, thus greatly improving the accuracy of weld identification. Advanced 3D image processing algorithms enable automatic and rapid weld identification, avoiding subjective errors and efficiency issues associated with manual identification.

[0028] Step S102: Based on the three-dimensional model, obtain the weld information corresponding to the current welding area. The weld information includes the shape, position and size of the weld.

[0029] In this embodiment, image recognition technology can be used to analyze the geometric features of the weld area in the generated 3D model, thereby accurately determining the location of the weld. Geometric features can include shape, curvature, etc. 3D measurement technology can be used to measure the dimensions of the weld, including width, depth, etc. Based on image recognition technology, the accuracy of the measurement results can be ensured, enabling precise ultrasonic scanning path planning in subsequent steps.

[0030] Step S103: Based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency, obtain the optimal ultrasonic scanning path corresponding to the current welding area. The movement parameters include the physical size and movement range of the ultrasonic scanning probe, and the scanning efficiency is a parameter characterizing the time required to complete the ultrasonic scanning task corresponding to the current welding area.

[0031] Determine the physical dimensions and range of motion of the ultrasonic scanning probe to ensure it can cover the entire weld area. Evaluate the scanning efficiency of different scanning paths, including scanning time and accuracy.

[0032] Path planning algorithms such as genetic algorithms and ant colony algorithms are used to optimize the scanning path, minimizing scanning time and improving scanning accuracy. By combining weld information, ultrasonic scanning probe parameters, and scanning efficiency, an optimal ultrasonic scanning path is generated, ensuring that the path fully covers the weld area and avoiding unnecessary repeated scanning.

[0033] Based on weld information, the physical parameters of the ultrasonic scanning probe, and scanning efficiency, an optimal ultrasonic scanning path is planned using an intelligent algorithm. This optimal ultrasonic scanning path ensures complete coverage of the weld area, avoiding missed detections due to probe limitations, while minimizing unnecessary scans, thus significantly improving detection efficiency.

[0034] Step S104: Based on the optimal ultrasonic scanning path, control the movement of the ultrasonic scanning probe, and based on the ultrasonic signal data collected in real time by the ultrasonic scanning probe, obtain the weld quality assessment information corresponding to the current welding area.

[0035] In this embodiment, the ultrasound scanning probe is controlled to move along the generated optimal ultrasound scanning path, ensuring stability during movement and acquiring ultrasound signal data in real time. The acquired ultrasound signal data is preprocessed (e.g., noise reduction, filtering) to improve the signal-to-noise ratio and clarity.

[0036] Feature extraction technology is used to extract key features related to weld quality from ultrasonic signal data. Based on the key features, weld quality assessment information corresponding to the current welding area is obtained. The weld quality assessment information may include information such as whether there are defects and the severity of the defects.

[0037] By controlling the ultrasonic scanning probe to move along an optimal path and acquiring ultrasonic signal data in real time, dynamic monitoring of weld quality is achieved. After processing and analysis, the acquired data can provide weld quality assessment information immediately. The real-time assessment function helps to promptly identify and address quality problems during the welding process, avoiding potential safety hazards in subsequent processing and use.

[0038] In this embodiment, before step S104, the method further includes: obtaining the initial operating parameters of the ultrasonic scanning probe based on the welding material corresponding to the current welding area, wherein the initial operating parameters include the scanning mode, frequency, and gain of the ultrasonic scanning probe.

[0039] Determine the welding material for the current welding area. Welding materials can include aluminum alloys, stainless steel, etc. The choice of welding material will directly affect the settings of parameters such as the scanning mode, frequency, and gain of the ultrasonic scanning probe.

[0040] Based on the characteristics of the welding material, a suitable scanning mode, frequency, and gain are selected from a preset scanning parameter library. Scanning modes can include pulse mode, continuous mode, etc., while the frequency and gain are adjusted according to the acoustic impedance and attenuation characteristics of the welding material. By setting appropriate initial operating parameters for the ultrasonic scanning probe based on the characteristics of the welding material, the targeting and effectiveness of the detection process can be ensured, improving the accuracy and reliability of the detection.

[0041] After step S104, the method further includes: acquiring environmental information corresponding to the current welding area in real time, and adjusting the initial working parameters based on the environmental information, wherein the environmental information includes temperature, humidity and noise.

[0042] Before or during the movement of the ultrasonic scanning probe, sensors are used to acquire real-time environmental information about the welding area, including temperature, humidity, and noise. Based on this environmental information, the initial operating parameters of the ultrasonic scanning probe are dynamically adjusted. For example, when the temperature rises, the gain can be increased to compensate for the attenuation of sound waves; when the humidity is high, the frequency can be adjusted to reduce the scattering of sound waves in the air.

[0043] By acquiring environmental information in real time and dynamically adjusting the operating parameters of the ultrasonic scanning probe, it is possible to adapt to the detection needs in different environments, thereby improving the adaptability and stability of the detection.

[0044] In this embodiment, step S103 specifically includes: obtaining the weld information, the movement parameters of the ultrasonic scanning probe, and the initial ultrasonic scanning path corresponding to the scanning efficiency based on a preset path generation algorithm; obtaining the weld quality prediction information of the current welding area based on the weld information; and adjusting the initial ultrasonic scanning path based on the weld quality prediction information to obtain the optimal ultrasonic scanning path.

[0045] Weld information can include key parameters such as the weld's start point, end point, width, and depth. Based on the weld's complexity and inspection requirements, movement parameters such as the ultrasonic scanning probe's moving speed, acceleration, and step distance are set. This ensures the probe can move smoothly and accurately along the weld while maintaining inspection efficiency. In this embodiment, complexity refers to a comprehensive consideration of the weld's geometry, structural features, material properties, quantity, and distribution.

[0046] Based on the urgency of the inspection task and the capabilities of the equipment, set a reasonable scanning speed and time. Balance inspection accuracy and efficiency to ensure high-quality inspection tasks are completed within a limited time.

[0047] The weld information, probe movement parameters, and scanning efficiency are input into a preset path generation algorithm. Based on the input information, the algorithm calculates an initial ultrasonic scanning path that covers the weld and meets the movement parameter and efficiency requirements.

[0048] A thorough analysis of weld information is conducted to extract key information such as weld geometry and material properties. These extracted weld features are then input into a weld quality prediction model. Based on historical data and machine learning algorithms, the model predicts weld quality, providing information such as defect type, location, and size.

[0049] Based on weld quality prediction information, critical areas within the weld that may contain defects are identified. These critical areas serve as the focus of inspection to ensure accurate defect detection. The initial ultrasonic scanning path is adjusted to increase the scanning density and accuracy of critical areas. Simultaneously, scanning of non-critical areas is reduced to improve inspection efficiency. The adjusted path should ensure comprehensive and accurate weld coverage while meeting inspection efficiency requirements.

[0050] In practical applications, the adjusted ultrasound scanning path can be validated. By comparing the detection results and prediction information, the effectiveness of the path adjustment can be evaluated. Based on the validation results, the path can be further optimized and improved.

[0051] By adjusting the ultrasonic scanning path based on weld information and quality prediction data, the inspection process can be more precisely targeted at critical areas of the weld, helping to reduce the risk of missed and false detections and improve the accuracy and reliability of the inspection. The optimized ultrasonic scanning path reduces unnecessary scans, improves inspection efficiency, helps shorten the inspection cycle, reduces inspection costs, and meets the fast-paced requirements of production lines. The ability to flexibly adjust the ultrasonic scanning path according to the characteristics and inspection requirements of different welds makes the ultrasonic scanning control method based on three-dimensional real-time imaging widely applicable, suitable for various types of friction welding quality inspection scenarios for energy storage fluid cooling plates.

[0052] In this embodiment, after step S103, the method further includes: importing the three-dimensional model into the created AR scene, automatically adjusting the position, size, and rotation angle of the three-dimensional model to match the real current welding area; annotating the weld information on the three-dimensional model in the AR scene, the annotation information including color information and texture information; drawing the preview line or animation corresponding to the optimal ultrasonic scanning path on the three-dimensional model in the AR scene; and updating the AR scene to the AR display device in real time based on the data synchronization mechanism to provide an interactive interface for the user based on the AR display device.

[0053] The 3D model is imported into a pre-created AR scene, and AR technology is used to integrate the model with the real environment. Utilizing the spatial positioning capabilities of AR technology, the position, size, and rotation angle of the 3D model are automatically adjusted to perfectly match the actual welding area.

[0054] On the 3D model within the AR scene, the location, shape, and defects of the weld are annotated based on weld information. Annotation information can include color and texture information to visually demonstrate the actual condition of the weld. Preview lines or animations corresponding to the optimal ultrasonic scanning path are then drawn on the 3D model of the AR scene. This helps users intuitively understand the scanning path, improving scanning accuracy and efficiency.

[0055] Establish a data synchronization mechanism to ensure that information in the AR scene can be updated in real time to reflect the latest weld information and scanning path.

[0056] Based on the established data synchronization mechanism, the updated AR scene is transmitted to the AR display device in real time. The AR display device can include AR glasses, tablets, etc. An interactive interface is provided on the AR display device, allowing users to interact with the AR scene through gestures, voice, etc., such as adjusting scanning parameters and viewing detailed weld information. By utilizing AR technology, 3D models, annotation information, and scan path previews are intuitively displayed to users, enhancing the visualization of the inspection process and helping users better understand and operate the system. By providing an intuitive interactive interface and real-time AR scene updates, the operational difficulty is reduced, enabling even non-professionals to easily perform welding quality inspection.

[0057] In this embodiment, step S104 includes: obtaining the weld quality assessment information based on the key features extracted from the ultrasonic signal data; if the weld quality assessment information indicates the presence of weld defects, obtaining the defect correction suggestions and defect cause analysis information corresponding to the weld defects, and generating a defect report based on the defect correction suggestions, defect cause analysis information, defect type, defect location, and defect size.

[0058] Based on the extracted key features, a trained deep learning model is used to evaluate the weld quality, yielding weld quality assessment information. This deep learning model can be trained and optimized based on historical data and expert experience to improve the accuracy of the assessment. The weld quality assessment information may include the weld's quality grade, the presence of defects, and the severity of those defects.

[0059] When weld quality assessment information indicates defects, the system automatically identifies the type of defect. Defect types can include different categories such as porosity, slag inclusions, and cracks. For different types of defects, corresponding defect correction suggestions are provided. For example, for porosity defects, it is recommended to increase the preheating temperature before welding or adjust welding parameters to reduce porosity. Simultaneously, the system analyzes the causes of the defects, such as impurities in the welding materials or improper welding procedures.

[0060] Based on information such as defect type, location, size, corrective suggestions, and root cause analysis, a defect report is automatically generated. The defect report can include a detailed description of the defect, image or video evidence, corrective suggestions, root cause analysis, and possible solutions. The defect report can be saved and transmitted as an electronic document for easy subsequent analysis and processing.

[0061] The defect report includes a detailed description of the defect, image or video evidence, corrective suggestions, and root cause analysis, providing strong support for subsequent repairs and improvements. By combining advanced technologies such as deep learning models, an intelligent solution for the friction welding quality inspection of energy storage liquid cooling plates is provided, contributing to the intelligent development of related industries.

[0062] In this embodiment, obtaining the weld quality assessment information based on the key features extracted from the ultrasonic signal data includes: adjusting the input features and output categories of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area, and constructing a target deep learning model based on the adjusted input features and output categories; extracting the key features based on the adjusted input features, and inputting the key features into the target deep learning model to obtain the weld quality assessment information output by the target deep learning model.

[0063] If the welding quality inspection of the current welding area is the first inspection of the overall inspection method, then the initial deep learning model can be a general deep learning model applicable to weld quality inspection tasks.

[0064] If the welding quality inspection of the current welding area is not the first inspection of the overall inspection method, the initial deep learning model can be the target deep learning model corresponding to the welding quality inspection process of the previous welding area.

[0065] Whether the welding quality inspection of the current welding area is the first inspection can be determined by checking system records, user input, or other methods.

[0066] Based on the welding material of the energy storage fluid cooling plate (e.g., aluminum alloy, stainless steel) and the welding process corresponding to the current welding area (e.g., friction stir welding), the input features and output categories of the initial deep learning model are adjusted. Input features can include time-domain features, frequency-domain features, and statistical features of the ultrasonic signal data, which can reflect the internal structure and material properties of the weld. The output category is set according to the needs of weld quality assessment, such as qualified, unqualified, and defect type. Defect types can include porosity, slag inclusions, cracks, etc.

[0067] Based on the adjusted input features and output category, a target deep learning model is constructed. The construction process may include model architecture design, parameter initialization, and loss function selection. The model is trained and optimized using ultrasonic signal data from the current or previous welding areas and weld quality assessment results. The model's accuracy and generalization ability can be improved by adjusting hyperparameters, using data augmentation techniques, and introducing regularization.

[0068] Based on the adjusted input features, key features are extracted from the ultrasonic signal data using signal processing techniques or feature extraction algorithms. These key features reflect the weld geometry, material uniformity, presence and characteristics of defects, etc. The extracted key features are then input into a trained target deep learning model for weld quality assessment. The target deep learning model infers from the input key features and outputs weld quality assessment information.

[0069] Depending on the welding material and welding process, the input features and output categories of the deep learning model can be flexibly adjusted to adapt to different inspection needs. Through training and optimization of the deep learning model, key features in ultrasonic signals can be accurately extracted, and weld quality can be accurately evaluated based on these features.

[0070] In this embodiment, before adjusting the input features and output category of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area, the method further includes: determining whether the welding material and welding process corresponding to the initial deep learning model are the same as the welding material and welding process corresponding to the current welding area; if they are different, then the step of adjusting the input features and output category of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area is performed; if they are the same, then the initial deep learning model is used as the target deep learning model corresponding to the current welding area.

[0071] In this embodiment, the welding material and welding process information of the current welding area can be obtained from records or user input. It is then checked whether the welding material and welding process corresponding to the initial deep learning model match the information of the current welding area.

[0072] If they are different, adjust the input features and output categories of the deep learning model according to the welding material and welding process of the current welding area.

[0073] If they are the same, the initial deep learning model can be used directly, that is, the initial deep learning model can be used directly as the target deep learning model corresponding to the current welding area. No additional model adjustment or training is required, and the weld quality can be evaluated directly.

[0074] By ensuring that the deep learning model matches the welding material and welding process of the current welding area, the accuracy of weld quality assessment can be significantly improved. When the welding material and welding process corresponding to the initial deep learning model are the same as the information of the current welding area, the existing deep learning model can be used directly without additional model adjustment or training, thus saving time and resources.

[0075] Based on the same technical concept, this application also provides an ultrasonic scanning control system based on three-dimensional real-time imaging, such as... Figure 2 As shown, the ultrasound scanning control system 200 based on three-dimensional real-time imaging mainly includes: The image capture module 201 is used to capture a three-dimensional image of the current welding area in real time based on the image acquisition device, and generate a three-dimensional model corresponding to the current welding area based on a preset image processing algorithm. The image acquisition device includes a 3D camera and a laser projector. The weld acquisition module 202 is used to acquire weld information corresponding to the current welding area based on the three-dimensional model. The weld information includes the shape, position and size of the weld. The path acquisition module 203 is used to acquire the optimal ultrasonic scanning path corresponding to the current welding area based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency. The movement parameters include the physical size and movement range of the ultrasonic scanning probe, and the scanning efficiency is a parameter that characterizes how fast it takes to complete the ultrasonic scanning task corresponding to the current welding area. The control movement module 204 is used to control the movement of the ultrasonic scanning probe based on the optimal ultrasonic scanning path, and to obtain weld quality assessment information corresponding to the current welding area based on the ultrasonic signal data collected in real time by the ultrasonic scanning probe.

[0076] Optionally, before controlling the movement module 204, the following is also included: The initial parameter acquisition module is used to acquire the initial operating parameters of the ultrasonic scanning probe based on the welding material corresponding to the current welding area. The initial operating parameters include the scanning mode, frequency, and gain of the ultrasonic scanning probe. Following the control of the movement module 204, the following is also included: The initial working parameter adjustment module is used to acquire environmental information corresponding to the current welding area in real time, and adjust the initial working parameters based on the environmental information, which includes temperature, humidity and noise.

[0077] Optionally, the path acquisition module 203 includes: The first acquisition submodule is used to acquire the weld information, the movement parameters of the ultrasonic scanning probe, and the initial ultrasonic scanning path corresponding to the scanning efficiency based on a preset path generation algorithm. The second acquisition submodule is used to acquire weld quality prediction information of the current welding area based on the weld information. The optimal path submodule is used to adjust the initial ultrasonic scanning path based on the weld quality prediction information to obtain the optimal ultrasonic scanning path.

[0078] Optionally, after obtaining the path module 203, the following is also included: The 3D import module is used to import the 3D model into the created AR scene and automatically adjust the position, size and rotation angle of the 3D model to match the real current welding area. The annotation module is used to annotate the annotation information corresponding to the weld information on the three-dimensional model in the AR scene, wherein the annotation information includes color information and texture information; and to draw the preview line or animation corresponding to the optimal ultrasonic scanning path on the three-dimensional model in the AR scene. The display interaction module is used to update the AR scene to the AR display device in real time based on the data synchronization mechanism, so as to provide an interactive interface for the user based on the AR display device.

[0079] Optionally, the control movement module 204 includes: The feature extraction submodule is used to obtain the weld quality assessment information based on the key features extracted from the ultrasonic signal data; The third acquisition submodule is used to acquire the defect correction suggestions and defect cause analysis information corresponding to the weld defect when the weld quality assessment information indicates that there is a weld defect, and generate a defect report based on the defect correction suggestions, defect cause analysis information, defect type, defect location and defect size.

[0080] Optionally, the feature extraction submodule includes: The model parameter adjustment submodule is used to adjust the input features and output categories of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area, and to construct the target deep learning model based on the adjusted input features and output categories. The fourth acquisition submodule is used to extract the key features based on the adjusted input features, and input the key features into the target deep learning model to obtain the weld quality assessment information output by the target deep learning model.

[0081] Optionally, before adjusting the model parameter submodule, the following is also included: The judgment submodule is used to determine whether the welding material and welding process corresponding to the initial deep learning model are the same as those corresponding to the current welding area; if they are different, the step of adjusting the input features and output category of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area is executed; if they are the same, the initial deep learning model is used as the target deep learning model corresponding to the current welding area.

[0082] In one example, a module in any of the above systems may be one or more integrated circuits configured to implement the above methods, such as one or more application-specific integrated circuits (ASICs), or one or more digital signal processors (DSPs), or one or more field-programmable gate arrays (FPGAs), or a combination of at least two of these integrated circuit forms.

[0083] For example, when modules in a system can be implemented through a processing element scheduler, the processing element can be a general-purpose processor, such as a central processing unit (CPU) or other processor capable of calling programs. Alternatively, these modules can be integrated together to form a system-on-a-chip (SOC).

[0084] In this application, various objects such as messages / information / devices / network elements / systems / apparatus / actions / operations / processes / concepts may be named. It is understood that these specific names do not constitute a limitation on the relevant objects. The names may be changed depending on the scenario, context, or usage habits. The understanding of the technical meaning of the technical terms in this application should be mainly determined from their functions and technical effects embodied / performed in the technical solution.

[0085] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0086] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0087] Based on the same technical concept, this application also provides an electronic device, such as... Figure 3 As shown, the electronic device 300 includes a processor 301 and a memory 302, and may further include one or more of an information input / output (I / O) interface 303, a communication component 304, and a communication bus 305.

[0088] The processor 301 controls the overall operation of the electronic device 300 to complete all or part of the steps in the aforementioned ultrasound scanning control method based on three-dimensional real-time imaging. The memory 302 stores various types of data to support the operation of the electronic device 300. This data may include, for example, instructions for any application or method operating on the electronic device 300, as well as application-related data. The memory 302 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as one or more of Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read-Only Memory (EPROM), Programmable Read-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0089] I / O interface 303 provides an interface between processor 301 and other interface modules, such as keyboards, mice, and buttons. These buttons can be virtual or physical. Communication component 304 is used to test wired or wireless communication between electronic device 300 and other devices. Wireless communication includes Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G, or 4G, or a combination thereof. Therefore, the corresponding communication component 304 may include a Wi-Fi component, a Bluetooth component, and an NFC component.

[0090] The communication bus 305 may include a path for transmitting information between the aforementioned components. The communication bus 305 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 305 can be divided into an address bus, a data bus, a control bus, etc.

[0091] The electronic device 300 may be implemented by one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to execute the ultrasonic scanning control method based on three-dimensional real-time imaging given in the above embodiments.

[0092] Electronic device 300 may include, but is not limited to, mobile terminals such as digital broadcast receivers, PDAs (personal digital assistants), and PMPs (portable multimedia players), as well as fixed terminals such as digital TVs and desktop computers, and may also be servers.

[0093] Based on the same technical concept, this application also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the above-described ultrasonic scanning control method based on three-dimensional real-time imaging.

[0094] The computer-readable storage medium may include various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0095] The terms “comprising,” “including,” or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0096] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this application, "multiple" means at least two, such as two, three, etc., unless otherwise explicitly specified.

[0097] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0098] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A method for controlling ultrasonic scanning based on three-dimensional real-time imaging, characterized in that, include: Based on an image acquisition device, a three-dimensional image of the current welding area is captured in real time, and a three-dimensional model corresponding to the current welding area is generated based on a preset image processing algorithm. The image acquisition device includes a 3D camera and a laser projector. Based on the three-dimensional model, the weld information corresponding to the current welding area is obtained, and the weld information includes the shape, position and size of the weld. Based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency, the optimal ultrasonic scanning path corresponding to the current welding area is obtained. The movement parameters include the physical size and range of motion of the ultrasonic scanning probe, and the scanning efficiency is a parameter that characterizes how fast the ultrasonic scanning task corresponding to the current welding area is completed. Based on the optimal ultrasonic scanning path, the ultrasonic scanning probe is moved, and based on the ultrasonic signal data collected in real time by the ultrasonic scanning probe, the weld quality assessment information corresponding to the current welding area is obtained.

2. The ultrasonic scanning control method based on three-dimensional real-time imaging according to claim 1, characterized in that, Before controlling the movement of the ultrasound scanning probe based on the optimal ultrasound scanning path, the method further includes: Based on the welding material corresponding to the current welding area, the initial operating parameters of the ultrasonic scanning probe are obtained, including the scanning mode, frequency, and gain of the ultrasonic scanning probe. After controlling the movement of the ultrasound scanning probe based on the optimal ultrasound scanning path, the method further includes: The system acquires environmental information corresponding to the current welding area in real time, and adjusts the initial working parameters based on the environmental information, which includes temperature, humidity, and noise.

3. The ultrasonic scanning control method based on three-dimensional real-time imaging according to claim 2, characterized in that, The step of obtaining the optimal ultrasonic scanning path corresponding to the current welding area based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency includes: Based on a preset path generation algorithm, the weld information, the movement parameters of the ultrasonic scanning probe, and the initial ultrasonic scanning path corresponding to the scanning efficiency are obtained. Based on the weld information, weld quality prediction information for the current welding area is obtained; Based on the weld quality prediction information, the initial ultrasonic scanning path is adjusted to obtain the optimal ultrasonic scanning path.

4. The ultrasonic scanning control method based on three-dimensional real-time imaging according to claim 3, characterized in that, After obtaining the optimal ultrasonic scanning path corresponding to the current welding area based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency, the method further includes: Import the 3D model into the created AR scene, and automatically adjust the position, size, and rotation angle of the 3D model to match the actual current welding area; On the 3D model in the AR scene, annotation information corresponding to the weld information is marked, including color information and texture information; and on the 3D model in the AR scene, a preview line or animation corresponding to the optimal ultrasonic scanning path is drawn. Based on the data synchronization mechanism, the AR scene is updated to the AR display device in real time, so as to provide an interactive interface for users based on the AR display device.

5. The ultrasonic scanning control method based on three-dimensional real-time imaging according to claim 2, characterized in that, The process of obtaining weld quality assessment information corresponding to the current welding area based on the ultrasonic signal data acquired in real time by the ultrasonic scanning probe includes: Based on the key features extracted from the ultrasonic signal data, the weld quality assessment information is obtained; If the weld quality assessment information indicates the presence of weld defects, then obtain the corresponding defect correction suggestions and defect cause analysis information, and generate a defect report based on the defect correction suggestions, defect cause analysis information, defect type, defect location, and defect size.

6. The ultrasonic scanning control method based on three-dimensional real-time imaging according to claim 5, characterized in that, The key features extracted from the ultrasonic signal data are used to obtain the weld quality assessment information, including: Based on the welding material and the welding process corresponding to the current welding area, the input features and output categories of the initial deep learning model are adjusted, and a target deep learning model is constructed based on the adjusted input features and output categories. Based on the adjusted input features, the key features are extracted and input into the target deep learning model to obtain the weld quality assessment information output by the target deep learning model.

7. The ultrasonic scanning control method based on three-dimensional real-time imaging according to claim 6, characterized in that, Before adjusting the input features and output category of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area, the method further includes: Determine whether the welding material and welding process corresponding to the initial deep learning model are the same as the welding material and welding process corresponding to the current welding area; If they are different, then the step of adjusting the input features and output category of the initial deep learning model based on the welding material and the welding process corresponding to the current welding area is executed. If they are the same, then the initial deep learning model is used as the target deep learning model corresponding to the current welding area.

8. An ultrasonic scanning control system based on three-dimensional real-time imaging, characterized in that, include: An image capture module is used to capture a three-dimensional image of the current welding area in real time based on an image acquisition device, and to generate a three-dimensional model corresponding to the current welding area based on a preset image processing algorithm. The image acquisition device includes a 3D camera and a laser projector. The weld acquisition module is used to acquire weld information corresponding to the current welding area based on the three-dimensional model. The weld information includes the shape, position, and size of the weld. The path acquisition module is used to acquire the optimal ultrasonic scanning path corresponding to the current welding area based on the weld information, the movement parameters of the ultrasonic scanning probe, and the scanning efficiency. The movement parameters include the physical size and movement range of the ultrasonic scanning probe, and the scanning efficiency is a parameter that characterizes how fast the ultrasonic scanning task corresponding to the current welding area is completed. The control and movement module is used to control the movement of the ultrasonic scanning probe based on the optimal ultrasonic scanning path, and to obtain weld quality assessment information corresponding to the current welding area based on the ultrasonic signal data collected in real time by the ultrasonic scanning probe.

9. An electronic device, characterized in that, It includes a processor and a memory, wherein the processor is coupled to the memory; The processor is configured to execute a computer program stored in the memory to cause the electronic device to perform the method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, It includes a computer program or instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1 to 7.

Images