Welding process visual real-time monitoring system and method based on DMD technology

Abstract

The invention discloses a welding process visual real-time monitoring system and method based on a DMD technology, and belongs to the field of welding monitoring, and the system comprises a visual collection module which is used for collecting an image of a welding area in real time; the DMD light supplementing control module is used for dynamically adjusting the light projected to the welding area according to the collected image; the image processing and analyzing module is used for preprocessing the collected image and extracting and analyzing the characteristics of the welding area; the welding process control and planning module is used for extracting a 3D contour of a welding area and planning welding process parameters according to the spatial structure light form change projected by the DMD light supplementing control module; and the control and feedback module is used for adjusting welding parameters in real time according to an analysis result of the image processing and analyzing module and giving an alarm when an abnormality is detected. Light supplement is accurately controlled through the DMD technology, the overexposure phenomenon caused by welding hard light is effectively restrained, a camera can clearly collect detailed information of a welding area, and the welding detection accuracy is improved.

Description

Technical Field

[0001] The present invention relates to the field of welding monitoring, and particularly to a visual real-time monitoring system and method for welding process based on DMD technology. Background Art

[0002] During the welding process, accurately monitoring the welding state is crucial for ensuring the welding quality. Traditional welding monitoring methods have many deficiencies, such as: 1. The strong light generated during the welding process is likely to cause overexposure of the camera imaging, making it difficult to clearly obtain the details of the welding area, and thus unable to accurately judge key information such as the quality of the weld seam and the state of the molten pool. As Figure 1 shows the imaging results of an ordinary active molten pool camera. It can be observed from the figure that there is a large amount of overexposure in the smooth molten pool area at the front end of the welding wire. This is because the surface of the molten pool in this area is smooth and prone to form beam reflection. For the active illumination molten pool monitoring scheme, there is overexposure in this area.

[0003] 2. The complex changes in welding ambient light, such as the difference between indoor and outdoor light and the interference of other light sources in the workshop, will affect the stability and accuracy of the monitoring system. Existing monitoring systems are difficult to adapt to these ambient light changes.

[0004] 3. For some minor welding defects, such as fine pores and cracks, the resolution and contrast of ordinary visual monitoring means are insufficient and it is difficult to effectively identify them. Summary of the Invention

[0005] The purpose of the present invention is to overcome the problems existing in the existing welding process monitoring, such as overexposure, poor adaptability to ambient light, and difficulty in identifying minor defects, and provide a visual real-time monitoring system and method for welding process based on DMD technology to improve the accuracy and reliability of welding process monitoring.

[0006] The purpose of the present invention is achieved through the following technical solutions: In the first aspect, a visual real-time monitoring system for welding process based on DMD technology is provided, including: A visual acquisition module: for real-time acquisition of images of the welding area; A DMD supplementary light control module: for dynamically adjusting the light projected onto the welding area according to the acquired images; An image processing and analysis module: for preprocessing the acquired images and extracting and analyzing the characteristics of the welding area; A welding process control and planning module: extracting the 3D contour of the welding area and planning welding process parameters according to the morphological changes of the spatial structured light projected by the DMD supplementary light control module; Control and feedback module: adjusts the welding process parameters in real time according to the analysis results of the image processing and analysis module, and issues an alarm when an abnormality is detected.

[0007] In some embodiments, the visual acquisition module includes a CMOS camera.

[0008] In some embodiments, the DMD fill light control module includes a DMD module controller, an array light source, a DMD spatial light modulator, and an outgoing light lens group connected in sequence; the DMD module controller is respectively connected to the array light source and the DMD spatial light modulator for control.

[0009] Second, a method for real-time visual monitoring of the welding process based on DMD technology is provided for a real-time visual monitoring system of the welding process based on DMD technology described in the first aspect, including the following steps: S1. Initialize the visual acquisition module, the image processing and analysis module, the welding process control and planning module, and the control and feedback module; S2. Real-time collect images of the welding area through the visual acquisition module; S3. According to the collected images, dynamically adjust the light projected onto the welding area through the DMD fill light control module; and trigger the visual acquisition module to perform image exposure and collect images of the welding area at this time; S4. Preprocess the images collected in step S3 through the image processing and analysis module, and extract and analyze the characteristics of the welding area; S5. According to the change in the spatial structured light form projected by the DMD fill light control module, extract the 3D contour of the welding area and plan the welding process parameters through the welding process control and planning module; S6. According to the analysis results of step S4, adjust the welding process parameters in real time through the control and feedback module, and issue an alarm when an abnormality is detected.

[0010] In some embodiments, it further includes the step of: S7. After the welding process ends, save the images and analysis data collected during the monitoring process.

[0011] In some embodiments, the dynamically adjusting the light projected onto the welding area through the DMD fill light control module includes: Scan the image and record the pixel positions of the overexposed areas in the image, input the pixel positions of the overexposed areas into the DMD module controller, and after the DMD module controller performs scale scaling and spatial position offset based on the pixel positions, output to control the corresponding DMD units in the DMD spatial light modulator to turn off the fill light.

[0012] In some embodiments, the image acquired in step S3 is preprocessed by the image processing and analysis module, including: The image is denoised and enhanced by an image processing algorithm.

[0013] In some embodiments, the features of the welding area are extracted and analyzed, including: Based on the region segmentation algorithm of the FCN fully convolutional neural network, the welding area is segmented and extracted.

[0014] In some embodiments, the 3D contour of the welding area is extracted by the welding process control and planning module, including: The sine grating image projected by the DMD fill light control module is used to calculate the principal value of the phase by the four-step phase-shifting method; According to the principal value of the phase, the 3D contour of the welding area is reconstructed.

[0015] In some embodiments, it further includes: According to the reconstructed 3D contour of the welding area, monocular weld bead tracking is performed.

[0016] It should be further noted that the technical features corresponding to the above embodiments can be combined or replaced with each other without conflict to form a new technical solution.

[0017] Compared with the prior art, the beneficial effects of the present invention are: 1. Effectively suppress overexposure: By precisely controlling the fill light through DMD technology, the light intensity of the welding area can be adjusted in real time, effectively suppressing the overexposure phenomenon caused by strong welding light, enabling the camera to clearly collect the detailed information of the welding area, and improving the accuracy of weld quality detection.

[0018] 2. Adapt to ambient light changes: The system can automatically adjust the DMD fill light parameters according to different welding ambient light conditions, ensuring that there is always appropriate lighting in the welding area, and improving the adaptability and stability of the monitoring system in complex environments.

[0019] 3. Improve the defect recognition ability: The multi-segment coded modulation light formed by DMD technology enhances the detailed features of the welding area. Combined with the image processing and analysis algorithm, it can more accurately identify tiny welding defects, such as fine pores and cracks, which helps to improve the welding quality.

[0020] 4. Realize 3D welding automatic navigation and tracking: The stripe structured light obtained by DMD spatial light modulation is used to extract the 3D dimensions of the entire welding area. Through the 3D topography reconstructed by the structured light, the direction of the weld bead at the front end of the molten pool can be clearly obtained, providing a navigation function for the welding of curved weld beads, enabling the welding robot to perform monocular weld bead tracking. Description of the Drawings

[0021] Figure 1 Schematic diagram of the imaging result of an existing ordinary active molten pool camera; Figure 2 Schematic diagram of a visual real-time monitoring system for the welding process based on DMD technology according to the present invention; Figure 3 Schematic diagram of the operation of the monitoring system according to the present invention; Figure 4 Schematic diagram of the structure of the DMD fill light control module according to the present invention; Figure 5 Flowchart of the monitoring method according to the present invention; Figure 6 Schematic diagram of overexposure control according to the present invention; Figure 7 Schematic diagram of the structure of the region segmentation algorithm according to the present invention; Figure 8 Schematic diagram of the region segmentation result according to the present invention; Figure 9 Flowchart of structured light multi-frame data acquisition according to the present invention; Figure 10 Schematic diagram of the 3D topography reconstruction result according to the present invention; Figure 11 Schematic diagram of weld bead navigation according to the present invention. Detailed implementation manners

[0022] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. The components of the embodiments of the present application described and illustrated herein can be arranged and designed in various different configurations. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts fall within the scope of protection of the present invention.

[0023] It should be noted that all the defects existing in the above prior art solutions are the results obtained by the inventor after practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed by the embodiments of the present application below for the above problems should be the contributions made by the inventor to the present application during the invention creation process, and should not be construed as the technical content known to those skilled in the art.

[0024] In response to the technical problems pointed out in the background art, the embodiments provided by the present invention are as follows: Referring to Figure 2 , in an exemplary embodiment, a visual real-time monitoring system for the welding process based on DMD technology includes: A visual acquisition module: for real-time acquisition of images of the welding area; DMD Fill Light Control Module: Used to dynamically adjust the light projected onto the welding area according to the acquired image; Image Processing and Analysis Module: Used to preprocess the acquired image and extract and analyze the characteristics of the welding area; Welding Process Control and Planning Module: According to the morphological changes of the spatial structured light projected by the DMD Fill Light Control Module, extract the 3D contour of the welding area and plan the welding process parameters; Control and Feedback Module: According to the analysis results of the Image Processing and Analysis Module, adjust the welding process parameters in real time and issue an alarm when an abnormality is detected.

[0025] Specifically, as Figures 3 - 4 shown, the visual acquisition module includes a CMOS camera for real-time acquisition of the molten pool image. The parameters of the camera are selected according to the requirements of the welding scenario, such as resolution, frame rate, etc., to ensure that the dynamic changes in the welding area can be clearly captured. The DMD Fill Light Control Module includes a DMD module controller (not marked in the figure) and an array light source, a DMD spatial light modulator, and an output light lens group connected in sequence; the DMD module controller is respectively connected to the array light source and the DMD spatial light modulator for control. The core component of the DMD Fill Light Control Module is the digital micromirror device (DMD), which consists of a large number of tiny mirrors. By controlling the angles of these mirrors, the light intensity, angle, and distribution projected onto the welding area can be precisely adjusted. The DMD Fill Light Control Module works in coordination with the visual acquisition module to adjust the fill light parameters in real time according to the acquired image information.

[0026] The Image Processing and Analysis Module processes and analyzes the image acquired by the visual acquisition module. First, preprocessing operations such as denoising and enhancement are performed on the image through image processing algorithms to improve the quality of the image. Then, image recognition technology is used to extract and analyze the characteristics of the welding area, such as the shape, position of the weld, the size and shape of the molten pool, the position and length of the welding wire, etc.

[0027] The Welding Process Control and Planning Module projects the spatial structured light formed by the DMD Fill Light Control Module onto the welding area, and extracts the target contours such as the weld and molten pool in the welding area through the morphological changes of the projected light, and plans and adjusts the welding parameters according to the size of the area to be filled.

[0028] The Control and Feedback Module controls and provides feedback on the welding process according to the results of the Image Processing and Analysis Module. If abnormalities such as weld deviation and defects are detected during the welding process, the Control and Feedback Module will send a signal to the welding equipment to adjust the welding parameters and at the same time issue an alarm to the operator.

[0029] Furthermore, the system is built as follows: 1. Select a suitable camera, such as an industrial camera with high resolution and high frame rate, to meet the monitoring requirements for the rapid dynamic changes during the welding process. Install the camera at a position where the welding area can be clearly photographed, and ensure that its field of view covers the entire welding area. The camera FOV parameters are calculated as follows:

[0030] Where HFOV is the horizontal field of view angle of the camera, WFOV is the vertical field of view angle of the camera, DFOV is the diagonal field of view angle of the camera, W is the physical width of the sensor CMOS, H is the physical length of the sensor CMOS, and f is the focal length of the lens.

[0031] 2. Select a DMD fill light control module with excellent performance. Determine the model and parameters of the DMD according to the size and light requirements of the welding scene. Synchronously control the DMD fill light control module with the camera to ensure the time matching of fill light and image acquisition.

[0032] 3. Build an image processing and analysis module. A computer or a dedicated image processing chip can be used, and the corresponding image processing and analysis software can be installed to realize the real-time processing and analysis of the acquired images.

[0033] 4. Establish a control and feedback module. Connect the output signal of the image processing and analysis module to the control interface of the welding equipment to realize the real-time adjustment and feedback of the welding process parameters. Specifically, the control module is connected to the actuator of the welding equipment through relevant motor control communication protocols and is responsible for the specific execution logic. The feedback module realizes the feedback closed-loop of the position, torque, etc. of the actuator through the detection of real-time encoder data and the current of the actuator motor, etc., to achieve more precise welding control. At the same time, the input information of the control module is provided by the image processing and analysis module, including the offset information between the welding wire and the weld bead, the position information between the welding wire and the molten pool, etc. The planning and execution functions of the control feedback module are realized through these information.

[0034] Furthermore, debug and optimize the system, specifically including: 1. After the system is built, conduct initialization debugging. Set parameters such as the exposure time and gain of the camera so that it can obtain clear images under different lighting conditions. At the same time, perform initialization settings on the DMD fill light control module to determine the initial fill light mode and parameters.

[0035] 2. Conduct welding experiments. Under different welding process conditions, collect images during the welding process and observe the working effect of the DMD fill light control module. Optimize and adjust the DMD fill light parameters according to the acquired image quality and analysis results, such as adjusting the angle of the mirror, changing the light intensity and distribution, etc., to achieve the best fill light effect.

[0036] 3. Optimize the image processing and analysis algorithms. By comparing the accuracy and efficiency of different algorithms in welding image analysis, select the most suitable algorithm and adjust its parameters to improve the recognition accuracy of the characteristics of the welding area.

[0037] 4. Test the response speed and accuracy of the control and feedback system to ensure that when welding abnormalities are detected, the welding parameters can be adjusted in a timely and accurate manner, and an alarm is sent to the operator.

[0038] In another exemplary embodiment, a visual real-time monitoring method for the welding process based on DMD technology is provided for the visual real-time monitoring system for the welding process based on DMD technology, referring to Figure 5 , including the following steps: S1. Perform initialization settings on the visual acquisition module, image processing and analysis module, welding process control and planning module, and control and feedback module; set the parameters of the camera, such as exposure time, gain, etc.; initialize the fill light mode and parameters of the DMD; load the image processing and analysis algorithms.

[0039] S2. Use the visual acquisition module to collect images of the welding area in real time at the set frame rate; S3. According to the collected images, dynamically adjust the light projected onto the welding area through the DMD fill light control module; and trigger the visual acquisition module to perform image exposure and collect the images of the welding area at this time; S4. Preprocess the images collected in step S3 through the image processing and analysis module, and extract and analyze the characteristics of the welding area through algorithms such as edge detection and feature extraction, such as calculating the center line of the weld, the area and shape of the molten pool, etc.

[0040] S5. According to the change in the spatial structured light form projected by the DMD fill light control module, extract the 3D contour of the welding area and plan the welding process parameters through the welding process control and planning module; specifically, calculate the weld width, the volume of the area to be filled, and the internal morphology of the weld bead at the current welding position through the geometric position relationship between the DMD projector and the camera, so as to output control parameters such as the height of the welding torch, welding speed, wire feeding speed, etc. corresponding to the current welding position.

[0041] S6. According to the analysis results of step S4, adjust the welding process parameters in real time through the control and feedback module and issue an alarm when an abnormality is detected. Specifically, judge whether the welding process is normal according to the results of the image processing and analysis module. If it is detected that the weld deviation exceeds the allowable range or a defect appears, the control and feedback module will send a control signal to the welding equipment to adjust parameters such as welding current, voltage, and welding speed, and at the same time send an alarm to the operator to prompt the problems that occur during the welding process.

[0042] Furthermore, it further includes the steps of: S7. After the welding process ends, the system stops working and saves the images and analysis data collected during the monitoring process for subsequent evaluation and traceability of the welding quality.

[0043] During supplementary lighting, the DMD supplementary lighting control module adjusts the state of the projection reflection surface. After the adjustment is completed, a DMDReadySignal signal is generated. After receiving this signal, the DMD module controller starts to trigger the supplementary lighting light source to generate a pulsed light beam. After a certain time delay, the vision acquisition module is triggered for image exposure, thereby obtaining a real-time welding monitoring image under the corresponding modulated light.

[0044] When adjusting the overexposed area of the image by DMD, the dynamic adjustment of the light projected onto the welding area by the DMD supplementary lighting control module includes: By capturing the molten pool image, scanning the image and recording the pixel positions of the overexposed areas in the image, inputting the pixel positions of the overexposed areas into the DMD module controller. After the DMD module controller performs scale scaling and spatial position offset based on the pixel positions, it outputs to control the corresponding DMD units in the DMD spatial light modulator to turn off the supplementary lighting. As Figure 6 shown, for the DMD units in the DMD spatial light modulator corresponding to the pixel positions of the overexposed areas of the image, when a certain position is overexposed, the right DMD unit on the right turns off the supplementary lighting, thereby avoiding pixel overexposure in the image space.

[0045] Furthermore, the preprocessing of the image collected in step S3 by the image processing and analysis module includes: Performing denoising and enhancement processing on the image through image processing algorithms (such as gray-scale transformation, filtering algorithms, etc.) to improve the clarity and contrast of the image.

[0046] Furthermore, as Figure 7 shown, based on the region segmentation algorithm of the FCN fully convolutional neural network, segmenting and extracting the welding area can effectively obtain the data of each part. The fully convolutional neural network can extract small targets, and the features of its small objects come from the features provided by the original input image. For larger regions of interest such as grooves, the algorithm restores the segmented image through downsampling pooling and upsampling, thereby realizing region segmentation. Figure 8 This is the result of the FCN network segmenting the molten pool image, where red represents the welding wire, blue represents the groove, green represents the weld bead, yellow represents the molten pool, and cyan represents the nozzle.

[0047] Furthermore, the extraction of the 3D contour of the welding area by the welding process control and planning module includes: Calculating the principal value of the phase using the four-step phase-shifting method for the sine grating image projected by the DMD supplementary lighting control module; Reconstruct the 3D profile of the welding area according to the principal value of the phase.

[0048] Specifically, the multi-frame data acquisition process of structured light is as Figure 9 shown. After a certain time delay, the vision acquisition module is triggered for image exposure. The time threshold of the delay is set to 5 ms. As long as the delay time is within 5 ms, a sinusoidal grating image is projected through the DMD device every 1 ms. In this embodiment, four sinusoidal grating images are obtained (i.e., i takes integers from 1 to 4), and the light intensity distribution can be expressed as: , where is the light intensity of the image at (x, y), is the background light intensity, is the modulation intensity of the fringe light, is the phase information. Through the four-step phase-shifting method, the phase shift value changes equally spaced four times between , that is, the projector projects four phase images with a phase difference of . For the rd encoded picture ( ), its encoding formula is:

[0049] Through these four images, the principal value of the phase can be solved using the formula:

[0050] This principal value of the phase is unique for each pixel point, thus marking the relative position on the projected image.

[0051] Furthermore, according to the relevant regions of image segmentation, including information such as the base metal surface, wire position, molten pool position, weld bead position, etc., the weld bead line can be obtained based on the junction position between the base metal surface and the weld bead. By obtaining the weld bead lines on both sides of the groove, the center tracking of the welding process can be carried out. Specifically, the internal parameters of the welding molten pool camera can be calibrated first to determine parameters such as its principal point position and the focal length of the imaging system. The imaging system formula is as follows:

[0052] After obtaining the camera calibration results, determine the scale factor between the physical space and the pixel space, combine the position of the welding wire and the pixel spacing between the two groove lines and convert it into the physical position, thus completing functions such as the swing control of the welding wire and the weld bead edge, the control of the welding wire length, and the control of the welding molten pool shape; when the position of the welding wire is closer to the left groove line, the welding torch can be controlled to move to the right, and vice versa. By identifying the welding wire, the actual length of the welding wire can be calculated through an appropriate scale factor to obtain the dry extension tracking data during the welding process, and at the same time, the geometric shape and size of the molten pool can be calculated to achieve the control of the welding process. Among them, when the wire penetration situation occurs during the welding process in the detected molten pool image area, by controlling the unified wire feeding speed of the welding process, the heat input and the molten iron filling amount during the welding process can be reduced, providing time for the solidification of the molten pool during the welding process to ensure that the welding wire can restart arc welding on the solidified metal substrate. At the same time, through the visual monitoring of the welding process, when the molten hole is small, the swing width can be appropriately controlled to increase, thereby reducing the risk of wire penetration and ensuring that the quality of the welding result meets the requirements. The control of the arc stability during the welding process can be achieved by automatically controlling the length of the welding wire during the welding process. When the welding material bulges, the welding wire will be melted shorter. After the visual detection of the shortening of the welding wire, the dry extension can be adjusted to keep the length of the welding wire stable, thereby maintaining the arc state.

[0053] Furthermore, the method further includes: Perform monocular weld bead tracking based on the 3D contour of the reconstructed welding area. Specifically, the welding robot can perform monocular weld bead tracking to provide a navigation function for the welding of curved weld beads. Through the 3D topography obtained by structured light reconstruction, the direction of the weld bead at the front end of the molten pool can be clearly obtained, as shown in Figure 10. By calculating the angle between the contour line of the feature points on the weld bead and the movement of the welding device, the current heading angle deviation of the welding mechanism can be obtained , and at the same time, the offset between the center position of the welding mechanism and the weld bead can also be determined through the center position of the current feature point , through , the tracking parameters of the welding mechanism to achieve the navigation function. As Figure 11 shown. Based on the results of the 3D reconstruction, the contour information of the current weld bead can be known. By integrating and calculating the volume of the area to be filled, the wire feeding speed of the welding power source, the welding current and voltage, the walking speed of the welding robot, etc. are set to ensure an appropriate filling amount:

[0054] Among them, is the radius of the welding wire, is the wire feeding speed, is the cladding efficiency, is the cross-sectional area of the groove, is the wire feeding speed, is the metal filling volume per unit time.

[0055] Taking the body welding in automobile manufacturing as an example, the visual real-time monitoring system of the present invention is installed in the body welding production line. During the welding process, the system collects the images of the welding area in real time. The DMD fill light control module effectively suppresses the overexposure phenomenon caused by the strong welding light, enabling the camera to clearly capture the shape of the weld seam and the state of the molten pool. The image processing and analysis module analyzes the collected images, accurately identifies the subtle defects in the weld seam, and feeds them back to the control and feedback system in a timely manner. The control and feedback system adjusts the welding parameters, optimizes the welding process, and improves the quality and production efficiency of body welding.

[0056] The above specific embodiments are detailed descriptions of the present invention. It cannot be determined that the specific embodiments of the present invention are only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention belongs, without departing from the concept of the present invention, several simple deductions and substitutions can still be made, which should all be regarded as belonging to the protection scope of the present invention.

Claims

1. A visual real-time monitoring system for welding process based on DMD technology, characterized in that, Including: Visual acquisition module: used to acquire images of the welding area in real time; DMD supplementary light control module: used to dynamically adjust the light projected onto the welding area according to the acquired images; Image processing and analysis module: used to preprocess the acquired images and extract and analyze the characteristics of the welding area; Welding process control and planning module: according to the morphological changes of the spatial structured light projected by the DMD supplementary light control module, extract the 3D contour of the welding area and plan the welding process parameters; Control and feedback module: according to the analysis results of the image processing and analysis module, adjust the welding process parameters in real time and issue an alarm when an abnormality is detected.

2. The visual real-time monitoring system for welding process based on DMD technology according to claim 1, characterized in that, The visual acquisition module includes a CMOS camera.

3. The visual real-time monitoring system for welding process based on DMD technology according to claim 1, characterized in that, The DMD supplementary light control module includes a DMD module controller, an array light source, a DMD spatial light modulator, and an outgoing light lens group connected in sequence; the DMD module controller is respectively connected to the array light source and the DMD spatial light modulator for control.

4. A visual real-time monitoring method for the welding process based on DMD technology, which is used for a visual real-time monitoring system for the welding process based on DMD technology described in any one of claims 1-3, and is characterized in that, Including the following steps: S1. Initialize the visual acquisition module, the image processing and analysis module, the welding process control and planning module, and the control and feedback module; S2. Acquire images of the welding area in real time through the visual acquisition module; S3. According to the acquired images, dynamically adjust the light projected onto the welding area through the DMD supplementary light control module; and trigger the visual acquisition module to perform image exposure and acquire the images of the welding area at this time; S4. Preprocess the images acquired in step S3 through the image processing and analysis module and extract and analyze the characteristics of the welding area; S5. According to the morphological changes of the spatial structured light projected by the DMD supplementary light control module, extract the 3D contour of the welding area and plan the welding process parameters through the welding process control and planning module; S6. According to the analysis results of step S4, adjust the welding process parameters in real time through the control and feedback module and issue an alarm when an abnormality is detected.

5. The visual real-time monitoring method for welding process based on DMD technology according to claim 4, wherein Also including the step: S7. After the welding process ends, save the images and analysis data acquired during the monitoring process.

6. The visual real-time monitoring method for welding process based on DMD technology according to claim 4, characterized in that The dynamically adjusting the light projected onto the welding area through the DMD supplementary light control module includes: Scanning the image and recording the pixel positions of the overexposed areas in the image, inputting the pixel positions of the overexposed areas into the DMD module controller, and after the DMD module controller performs scale scaling and spatial position offset based on the pixel positions, output to control the corresponding DMD units in the DMD spatial light modulator to turn off the supplementary light.

7. A visual real-time monitoring method for welding process based on DMD technology according to claim 4, characterized in that The preprocessing the images acquired in step S3 through the image processing and analysis module includes: Performing denoising and enhancement processing on the images through image processing algorithms.

8. A visual real-time monitoring method for welding process based on DMD technology according to claim 4, characterized in that, The extracting and analyzing the characteristics of the welding area includes: Performing segmentation extraction on the welding area based on the region segmentation algorithm of the FCN fully convolutional neural network.

9. A visual real-time monitoring method for welding process based on DMD technology according to claim 4, characterized in that, The extracting the 3D contour of the welding area through the welding process control and planning module includes: Calculating the principal value of the phase by using the four-step phase-shifting method for the sine grating image projected by the DMD supplementary light control module; Reconstructing the 3D contour of the welding area according to the principal value of the phase.

10. The visual real-time monitoring method for welding process based on DMD technology according to claim 9, wherein, Also including: Performing monocular weld bead tracking according to the reconstructed 3D contour of the welding area.

Images