Welding arc detection method based on pulse camera

Abstract

The invention discloses a welding arc detection method based on a pulse camera, which comprises the following steps of: performing dynamic scene capture by using the pulse camera in an arc welding working scene, and outputting a corresponding pulse data stream; based on the pulse data flow, density enhancement, multi-scale integration and fusion reconstruction are carried out in sequence, and a high-quality reconstructed arc image is obtained; analyzing and processing the high-quality reconstructed arc image to obtain an arc geometric parameter and a temperature field parameter; quantitative evaluation of the welding arc stability and optimization of the welding quality are achieved according to the arc parameters. According to the method, the pulse camera with the time statistical characteristic is applied, the arc parameters reflecting the dynamic characteristics of the arc are captured in the arc welding work, support is provided for welding state stability monitoring and welding quality detection, and defects and abnormities which are difficult to capture by a traditional camera are helped to be monitored.

Description

Technical Field

[0001] This invention relates to the field of intelligent welding technology, and in particular to a welding arc detection method based on a pulse camera. Background Technology

[0002] In modern industrial manufacturing, welding, as a critical joining process, directly impacts the safety and durability of products. Traditional welding quality inspection methods, limited by efficiency and accuracy, have been gradually replaced by advanced vision-based monitoring technologies. Among these, pulse cameras, leveraging their ultra-high-speed imaging capabilities, provide a highly efficient means of monitoring and controlling welding quality.

[0003] Traditional welding arc detection methods suffer from poor real-time performance, low accuracy, and reliance on human experience, making it difficult to comprehensively and accurately assess arc stability. Furthermore, traditional sensors are easily limited in strong light and high-temperature environments. Existing technologies, such as the SSIR framework based on deep spiking neural networks, emphasize low energy consumption and temporal processing, but suffer from single-path integration limitations and cannot flexibly handle variable-density pulses. The HSSR framework improves high-speed object recognition through feature correlation of long and short-time pulse streams, but focuses on short-time feature extraction. Hierarchical temporal filtering frameworks utilize temporal correlations on motion trajectories to suppress noise and blur, achieving clear reconstruction through short-time and long-time filtering, but suffer from problems such as reliance on supervision or the complexity of motion estimation. This invention proposes a multi-scale strategy that balances global and local receptive fields by processing different scale window lengths in parallel, dynamically adapting to variable-density pulse signals, significantly suppressing noise and improving reconstruction accuracy; reducing computational cost and complexity, achieving efficient data compression and real-time processing; and integrating cross-scale correlations to achieve global balance while preserving local details, improving image quality and application robustness. With its high-speed imaging and strong anti-interference capabilities, pulse camera technology can capture arc morphology and temperature distribution in real time, providing more accurate arc stability analysis and overcoming the limitations of traditional methods. It is expected to become an important means of welding quality monitoring and promote the optimization and improvement of welding processes.

[0004] Pulse camera technology offers a new perspective in welding arc inspection, providing clear images even in intense arc light and complex welding environments. It captures transient phenomena during the welding process through high-speed imaging, enabling operators to accurately monitor weld quality. The arc cross-sectional geometry recorded by the pulse camera, along with the calculated temperature field distribution, can be used to detect arc stability, thereby controlling weld quality. Summary of the Invention

[0005] To address the problems existing in the prior art, this invention proposes a welding arc detection method based on a pulse camera. By introducing pulse camera technology and utilizing its ultra-high-speed imaging characteristics, transient phenomena in the welding process can be captured to accurately monitor welding quality.

[0006] To achieve the above-mentioned technical objectives, the present invention provides the following technical solution: A welding arc detection method based on a pulse camera specifically includes the following steps: S1. In the arc welding scenario, a pulse camera is introduced to capture the pulse data stream of the arc changing rapidly during the welding process; S2. Perform pre-integration processing on the pulse data stream obtained in step S1 to obtain a density-enhanced pulse signal; S3. Perform multi-scale path integration on the pulse data stream after pre-integration processing to obtain the multi-path integration result; S4. The multipath integration results are fused and reconstructed to obtain a high-quality reconstructed arc image; S5. First, perform threshold segmentation and image filling operations on the high-quality arc reconstruction image; S6. Further perform arc edge detection and geometric parameter extraction on the segmented image; the geometric parameters include the arc area and perimeter. S7. Extract temperature field parameters from the arc image after edge detection; S8. Combine the obtained arc geometric parameters and temperature field parameters to perform arc stability analysis.

[0007] Furthermore, step S2 specifically includes: S21. Input raw binary pulse data stream ,in For pixel coordinates, For time indexing; calculate the preliminary intensity estimate, expressed by the formula: ; in, It is the trigger threshold. It is the pulse interval; S22. Use a 3×3 convolution kernel to process the initial intensity estimation and output the density map. The formula is expressed as: ; in, It represents the neighborhood; m represents the neighborhood. Each element in; These are learnable weights; S23, Density Map By applying the ReLU activation function and ensuring that all values ​​are non-negative, a density-enhanced pulse signal is obtained. The formula is expressed as: ; The size is the same as the input, and it is used for subsequent multi-scale integration.

[0008] Furthermore, step S3 specifically includes: S31, Input density enhancement pulse signal Multiple independent paths are used, each corresponding to a different integration scale; S32. For each path, calculate the adaptive kernel. The formula is expressed as: ; ; in, It is global density. , Path-specific parameters are learned through backpropagation. For density plots, For pixel coordinates, For time indexing; The integral result for each path for: ; Where j is the path index; `i` is the window size for the `j`-th path; `i` is the traversal index, representing the relative time offset within the time window, ranging from 1 to `i`. ; S33. After each integration path, add a 1×1 convolution to fuse the spatial neighborhood, and use the Softplus activation function to ensure smooth accumulation, outputting the final multi-path smooth integration result. Each integration result has a size of H×W×T, where H, W, and T represent the image height, image width, and time step, respectively; different accumulation strategies are explored in parallel to ensure adaptation to high and low density pulses.

[0009] Furthermore, step S4 specifically includes: S41. Input the multipath smoothing integral result. Calculate the strength variance for each path. The formula is expressed as: ; in, The average intensity; For pixel coordinates, For time indexing; S42. Select the path with the lowest variance as the main path, and then weight and merge the other paths. The formula is as follows: ; in, For weighted fused images; It is the weighting coefficient for the j-th path, and its calculation formula is: ; in, It is an exponentially decaying function. It is a learnable scaling parameter used to control the sensitivity of the weights; S43. Apply 1×1 convolution to fuse channel information, then use the Sigmoid function for normalization to ensure that the output intensity value is within the range of [0,1], avoiding overflow and enhancing nonlinear mapping, thereby generating a high-quality reconstructed arc image with low noise and high consistency. The formula is expressed as: ; ; in, b is the convolution weight, c is the bias, and 'c' is the channel index. Information is integrated into the channel.

[0010] Furthermore, step S5 specifically includes: S51, Threshold segmentation; Set grayscale threshold. The arc region is segmented from the background to obtain a binarized image. ; S52, Image Filling: The closing operation is used to fill the binarized image to obtain a complete filled image. .

[0011] Furthermore, step S6 specifically includes: S61. Apply the Sobel operator for edge detection, and find the edge location by calculating the gradient and gradient magnitude of the image; S62. After accurately identifying the edges, the geometric parameters of the arc are extracted by analyzing the edge information and applying geometric calculation methods. The geometric parameters include: the longitudinal section of the arc column, the perimeter, and the area, where the area is calculated by the number of foreground pixels in the binarized image; the perimeter is calculated by the number of edge pixels obtained by edge detection.

[0012] Furthermore, step S7 specifically includes: S71. Calculate the projected radiation intensity. Assuming the welding arc is axisymmetrically distributed, extract the side projection intensity from the arc profile obtained by edge detection. The distribution of , where r is the radial distance from the arc axis to the edge; projected intensity The line integral radiation intensity, obtained by averaging the image intensity along the arc axis, is expressed by the formula: ; in, It is the length of the electric arc axis. Refers to time index Below, along the axis coordinates The image intensity value at the radial coordinate r perpendicular to the axis; S72. Calculate radial emissivity using the inverse Abel transform, and then use the inverse Abel transform to calculate the radial emissivity from the projected intensity. Restore radial emissivity It represents the volume emissivity of the electric arc at a radial position r, expressed by the formula: ; in, It is the integral variable; the upper limit of integration ∞ is replaced by the arc radius R in actual calculations, and the radial emissivity at each discrete point is obtained by discretization using numerical integration methods. : ; in, p represents the discrete point index; is the radial step size, and N is the number of discrete points; S73. Calculate the radial temperature distribution: Based on the Stefan-Boltzmann law, the radial emissivity... Convert to radial temperature Assuming the electric arc is a blackbody or a graybody, then The calculation formula is: ; in, It is the Stefan-Boltzmann constant, and the temperature unit is Kelvin.

[0013] Furthermore, step S8 specifically includes: S81. Using the area and perimeter of consecutive frame images, calculate the temporal change rate of geometric features to quantify the stability of the arc shape; the temporal change rate of geometric features includes the area change rate. and the rate of change of perimeter , defined as the relative difference between the current frame and the previous frame, is expressed by the formula: ; in, It is the arc area in frame t, in pixels². It is the arc area of ​​the previous frame; It is the arc circumference in frame t, in pixels; It represents the arc circumference of the previous frame; |·| represents the absolute value; S82. Calculate the temperature field uniformity; based on radial temperature. Calculate the standard deviation of temperature and temperature gradient This is used to quantify the stability of heat distribution; the formula is expressed as: ; ; Where N is the number of radial discrete points; It is the first p Radial temperature at discrete points; It is the average radial temperature. ; Radial step size; S83, Quantitative Stability Index; Calculate the overall stability index by combining geometric and temperature characteristics. To assess arc stability; the formula is expressed as: ; in, It is the average area change rate across multiple frames; It is the average perimeter change rate across multiple frames; , , It is a weighting coefficient; it sets the arc stability threshold to determine whether welding parameters need to be adjusted.

[0014] Based on the above technical solution, the present invention has the following beneficial effects: The method proposed in this invention introduces pulse camera technology, combining its ultra-high-speed imaging characteristics to capture pulse data of the welding arc during the arc welding process. After multi-scale integral reconstruction processing, a clear image is obtained. Following a series of processes including edge detection, image optimization, feature extraction, and temperature field calculation, the final output is the geometric parameters of the arc and the temperature field distribution. These results can intuitively reflect the dynamic characteristics of the arc during the welding process, providing important data support for welding quality control, process optimization, and equipment design. It can also monitor defects and anomalies that are difficult to capture with traditional cameras. Attached Figure Description

[0015] Figure 1 This is an overall flowchart of a welding arc detection method based on a pulse camera proposed in this invention; Figure 2 This is a schematic diagram illustrating the specific process of the method proposed in this invention. Detailed Implementation

[0016] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0017] Although the steps in this invention are arranged by reference numerals, this is not intended to limit the order of the steps. Unless the order of the steps is explicitly stated or the execution of a step requires other steps as a basis, the relative order of the steps can be adjusted. It is understood that the term "and / or" as used herein refers to and covers any and all possible combinations of one or more of the associated listed items.

[0018] like Figure 1-2 As shown, this invention proposes a welding arc detection method based on a pulse camera, which specifically includes the following steps: S1. In the arc welding scenario, a pulse camera is introduced to capture the pulse data stream of the arc changing rapidly during the welding process; these high-speed pulse images serve as the basis for in-depth analysis of the stability of the welding process.

[0019] S2. Perform pre-integration processing on the pulse data stream obtained in step S1 to obtain a density-enhanced pulse signal; In a preferred embodiment, step S2 specifically includes: S21. Input raw binary pulse data stream ,in For pixel coordinates, For time indexing; calculate the preliminary intensity estimate, expressed by the formula: ; in, It is the trigger threshold. It is the pulse interval; this step generates a noisy initial image; S22. Use a 3×3 convolution kernel to process the initial intensity estimation and output the density map. The formula is expressed as: ; in, It represents the neighborhood; m represents the neighborhood. Each element in; These are learnable weights; S23, Density Map By applying the ReLU activation function and ensuring that all values ​​are non-negative, a density-enhanced pulse signal is obtained. The formula is expressed as: ; The size is the same as the input, and it is used for subsequent multi-scale integration.

[0020] This embodiment reduces the variance of the overall image data by suppressing the initial quantization noise generated in step S21 through density smoothing.

[0021] S3. Perform multi-scale path integration on the pulse data stream after pre-integration processing to obtain the multi-path integration result; In a preferred embodiment, step S3 specifically includes: S31, Input density enhancement pulse signal The invention employs multiple independent paths, each corresponding to a different integration scale (accumulating signals from several consecutive time steps over a time window of different lengths to calculate instantaneous light intensity, with each time step corresponding to one sampling period of the camera). In this invention, three independent paths are used: a 50-step integration scale for a short time window, a 20-step integration scale for a medium time window, and a 50-step integration scale for a long time window. In this embodiment, multiple paths are designed for integration to improve robustness and adaptability: by processing window lengths of different scales in parallel, the global and local receptive fields are balanced, dynamically adapting to variable density pulse signals (such as strong light interference from welding arcs), significantly suppressing noise and improving reconstruction accuracy; on the other hand, computational costs and complexity are reduced: multi-scale parallel fusion expands the receptive field while maintaining linear complexity, avoiding the limitations of single scales, and achieving efficient data compression and real-time processing; at the same time, detail preservation and balance are enhanced: cross-scale correlations (such as spatiotemporal aggregation) are integrated to achieve global balance (such as temperature field uniformity) while preserving local details, improving image quality and application robustness; S32. For each path, calculate the adaptive kernel. The formula is expressed as: ; ; in, It is global density. , Path-specific parameters are learned through backpropagation. For density plots, For pixel coordinates, For time indexing; The integral result for each path for: ; Where j is the path index; `i` is the window size for the `j`-th path; `i` is the traversal index, representing the relative time offset within the time window, ranging from 1 to `i`. ; S33. After each integration path, add a 1×1 convolution to fuse the spatial neighborhood, and use the Softplus activation function to ensure smooth accumulation, outputting the final multi-path smooth integration result. Each integration result has a size of H×W×T (H, W, and T represent the image height, image width, and time step, respectively); different accumulation strategies are explored in parallel to ensure adaptation to high and low density pulses.

[0022] S4. The multipath integration results are fused and reconstructed to obtain a high-quality reconstructed arc image; In a preferred embodiment, step S4 specifically includes: S41. Input the multipath smoothing integral result. Calculate the strength variance for each path. The formula is expressed as: ; in, The average intensity; For pixel coordinates, For time indexing; S42. Select the path with the lowest variance as the main path, and then weight and merge the other paths. The formula is as follows: ; in, For weighted fused images; It is the weighting coefficient for the j-th path, and its calculation formula is: ; in, It is an exponentially decaying function. It is a learnable scaling parameter used to control the sensitivity of the weights.

[0023] In this embodiment, the best integration result is selected and fused from the multi-path integration results to achieve soft decision fusion, ensuring that the final output integrates the advantages of multiple paths, while prioritizing low-noise paths to improve the temporal consistency and overall quality of reconstruction.

[0024] S43. Apply 1×1 convolution to fuse channel information, then use the Sigmoid function for normalization to ensure that the output intensity value is within the range of [0,1], avoiding overflow and enhancing nonlinear mapping, thereby generating a high-quality reconstructed arc image with low noise and high consistency. The formula is expressed as: ; ; in, b is the convolution weight, c is the bias, and 'c' is the channel index. Information is integrated into the channel.

[0025] This ensures The output intensity values ​​are in the range of [0,1], avoiding overflow and enhancing nonlinear mapping, generating a final image sequence with low noise and high consistency.

[0026] S5. First, perform threshold segmentation and image filling operations on the high-quality arc reconstruction image; In a preferred embodiment, step S5 specifically includes: S51, Threshold segmentation; Set grayscale threshold. The arc region is segmented from the background to obtain a binarized image. The formula is expressed as: ; in, Coordinates in the image The corresponding grayscale value; S52, Image Filling: The closing operation is used to fill the binarized image to obtain a complete filled image. The formula is expressed as: ; in, It is a structural element; For expansion operation, This is a corrosion operation.

[0027] In this embodiment, step S5 directly applies the high-quality image reconstructed in step S4 to generate a binarized and complete arc region image, which is used as subsequent input. At the same time, threshold segmentation and image filling ensure the removal of noise and interference, simplification of data structure, and improvement of accuracy and robustness.

[0028] S6. Further perform arc edge detection and geometric parameter extraction on the segmented image; the geometric parameters include the arc area and perimeter. In a preferred embodiment, step S6 specifically includes: S61. Apply the Sobel operator for edge detection, finding edge locations by calculating the image gradient and gradient magnitude; the Sobel edge detection calculation formula is: ; ; in, and These represent the images in and Gradient of direction; Gradient magnitude of edge intensity The calculation formula is: ; In this embodiment, by calculating the gradient and gradient magnitude of the image, the Sobel operator can help find areas in the image where the grayscale values ​​change significantly, which may be the edges of objects. Since the welding arc itself usually has a large grayscale difference (higher or lower brightness) from the background, using the Sobel operator to reflect this significant change can effectively identify the edges of the arc.

[0029] S62. After accurately identifying the edges, the geometric parameters of the arc are extracted by analyzing the edge information and applying geometric calculation methods. The geometric parameters include: the longitudinal section of the arc column, its perimeter, and its area. The area is calculated using the number of foreground pixels in the binarized image; the perimeter is calculated using the number of edge pixels obtained from edge detection. The specific calculation process is as follows: area A The formula for calculating the number of foreground pixels in a binarized image is as follows: ; in, This represents the pixel value in the binary image. If it is 1, it means that the pixel belongs to the arc region; if it is 0, it does not. perimeter C The number of edge pixels obtained through edge detection is calculated using the following formula: ; in, This represents the pixels that indicate the edge of the electric arc in the edge detection results.

[0030] S7. Extract temperature field parameters from the arc image after edge detection; In a preferred embodiment, step S7 specifically includes: S71. Calculate the projected radiation intensity. Assuming the welding arc is axisymmetrically distributed, extract the side projection intensity from the arc profile obtained by edge detection. The distribution of , where r is the radial distance from the arc axis to the edge; projected intensity The line integral radiation intensity, obtained by averaging the image intensity along the arc axis, is expressed by the formula: ; in, It is the length of the electric arc axis. Refers to time index Below, along the axis coordinates The image intensity value at the radial coordinate r perpendicular to the axis; S72. Calculate radial emissivity using the inverse Abel transform, and then use the inverse Abel transform to calculate the radial emissivity from the projected intensity. Restore radial emissivity It represents the volume emissivity of the electric arc at a radial position r, expressed by the formula: ; in, It is the integral variable; the upper limit of integration ∞ is replaced by the arc radius R in actual calculations, and the radial emissivity at each discrete point is obtained by discretization using numerical integration methods. : ; in, p represents the discrete point index, which takes a value from 0 to N-1; is the radial step size, and N is the number of discrete points; S73. Calculate the radial temperature distribution: Based on the Stefan-Boltzmann law, the radial emissivity... Convert to radial temperature Assuming the electric arc is a blackbody or a graybody, then The calculation formula is: ; in, It is the Stefan-Boltzmann constant, and the temperature unit is Kelvin.

[0031] S8. Combine the obtained arc geometric parameters and temperature field parameters to perform arc stability analysis.

[0032] In a preferred embodiment, step S8 specifically includes: S81. Using the area and perimeter of consecutive frame images, calculate the temporal change rate of geometric features to quantify the stability of the arc shape; the temporal change rate of geometric features includes the area change rate. and the rate of change of perimeter , defined as the relative difference between the current frame and the previous frame, is expressed by the formula: ; in, It is the arc area in frame t, in pixels². It is the arc area of ​​the previous frame; It is the arc circumference in frame t, in pixels; It represents the arc circumference of the previous frame; |·| represents the absolute value; this step evaluates shape fluctuations, and the smaller the rate of change, the more stable the arc geometry, which is suitable for detecting flickering or offset.

[0033] S82. Calculate the temperature field uniformity; based on radial temperature. Calculate the standard deviation of temperature and temperature gradient This is used to quantify the stability of heat distribution; the formula is expressed as: ; ; Where N is the number of radial discrete points; It is the first p Radial temperature at discrete points; It is the average radial temperature. ; The radial step size; this step assesses thermal non-uniformity. smaller or A smooth temperature field indicates a more stable temperature field, which helps avoid defects caused by overheating or cold spots.

[0034] S83, Quantitative Stability Index; Calculate the overall stability index by combining geometric and temperature characteristics. To assess arc stability; the formula is expressed as: ; in, It is the average area change rate across multiple frames; It is the average perimeter change rate across multiple frames; , , It is a weighting coefficient; it sets the arc stability threshold to determine whether welding parameters need to be adjusted. The range is [0,1], with values ​​close to 1 indicating high stability. In this embodiment, a threshold is set for judgment: if... If it is unstable, it will be marked as unstable, triggering welding parameter adjustment.

[0035] In summary, the present invention proposes a welding arc detection method based on a pulse camera. By introducing a pulse camera with time statistical characteristics, it captures arc parameters reflecting the dynamic characteristics of the arc during arc welding, providing support for state stability monitoring and welding quality inspection, and helping to monitor defects and anomalies that are difficult to capture by traditional cameras.

[0036] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of this application. 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 those different embodiments or examples.

[0037] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus or device (such as a computer-based system, a processor-included system or other system that can fetch and execute instructions from, an instruction execution system, apparatus or device).

[0038] The above embodiments provide a detailed description of the present invention. Specific examples have been used to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of the present invention. Therefore, the content of this specification should not be construed as a limitation of the present invention.

Claims

1. A method for detecting welding arcs based on a pulse camera, characterized in that, Includes the following steps: S1. In the arc welding scenario, a pulse camera is introduced to capture the pulse data stream of the arc changing rapidly during the welding process; S2. Perform pre-integration processing on the pulse data stream obtained in step S1 to obtain a density-enhanced pulse signal; S3. Perform multi-scale path integration on the pulse data stream after pre-integration processing to obtain the multi-path integration result; S4. The multipath integration results are fused and reconstructed to obtain a high-quality reconstructed arc image; S5. First, perform threshold segmentation and image filling operations on the high-quality arc reconstruction image; S6. Further perform arc edge detection and geometric parameter extraction on the segmented image; the geometric parameters include the arc area and perimeter. S7. Extract temperature field parameters from the arc image after edge detection; S8. Combine the obtained arc geometric parameters and temperature field parameters to perform arc stability analysis.

2. The welding arc detection method based on a pulse camera according to claim 1, characterized in that, Step S2 specifically includes: S21. Input raw binary pulse data stream ,in For pixel coordinates, For time indexing; calculate the preliminary intensity estimate, expressed by the formula: ； in, It is the trigger threshold. It is the pulse interval; S22. Use a 3×3 convolution kernel to process the initial intensity estimation and output the density map. The formula is expressed as: ； in, It represents the neighborhood; m represents the neighborhood. Each element in; These are learnable weights; S23, Density Map By applying the ReLU activation function and ensuring that all values ​​are non-negative, a density-enhanced pulse signal is obtained. The formula is expressed as: ； The size is the same as the input, and it is used for subsequent multi-scale integration.

3. The welding arc detection method based on a pulse camera according to claim 1, characterized in that, Step S3 specifically includes: S31, Input density enhancement pulse signal Multiple independent paths are used, each corresponding to a different integration scale; S32. For each path, calculate the adaptive kernel. The formula is expressed as: ； ； in, It is global density. , Path-specific parameters are learned through backpropagation. For density plots, For pixel coordinates, For time indexing; The integral result for each path for: ； Where j is the path index; `i` is the window size for the `j`-th path; `i` is the traversal index, representing the relative time offset within the time window, ranging from 1 to `i`. ; S33. After each integration path, add a 1×1 convolution to fuse the spatial neighborhood, and use the Softplus activation function to ensure smooth accumulation, outputting the final multi-path smooth integration result. Each integration result has a size of H×W×T, where H, W, and T represent the image height, image width, and time step, respectively; different accumulation strategies are explored in parallel to ensure adaptation to high and low density pulses.

4. The welding arc detection method based on a pulse camera according to claim 1, characterized in that, Step S4 specifically includes: S41. Input the multipath smoothing integral result. Calculate the strength variance for each path. The formula is expressed as: ； in, The average intensity; For pixel coordinates, For time indexing; S42. Select the path with the lowest variance as the main path, and then weight and merge the other paths. The formula is as follows: ； in, For weighted fused images; It is the weighting coefficient for the j-th path, and its calculation formula is: ； in, It is an exponentially decaying function. It is a learnable scaling parameter used to control the sensitivity of the weights; S43. Apply 1×1 convolution to fuse channel information, then use the Sigmoid function for normalization to ensure that the output intensity value is within the range of [0,1], generating a high-quality reconstructed arc image with low noise and high consistency. The formula is expressed as: ； ； in, b is the convolution weight, c is the bias, and 'c' is the channel index. Information is integrated into the channel.

5. The welding arc detection method based on a pulse camera according to claim 1, characterized in that, Step S5 specifically includes: S51, Threshold segmentation; Set grayscale threshold. The arc region is segmented from the background to obtain a binarized image. ; S52, Image Filling: The closing operation is used to fill the binarized image to obtain a complete filled image. .

6. The welding arc detection method based on a pulse camera according to claim 1, characterized in that, Step S6 specifically includes: S61. Apply the Sobel operator for edge detection, and find the edge location by calculating the gradient and gradient magnitude of the image; S62. After accurately identifying the edges, the geometric parameters of the arc are extracted by analyzing the edge information and applying geometric calculation methods. The geometric parameters include: the longitudinal section of the arc column, the perimeter, and the area, where the area is calculated by the number of foreground pixels in the binarized image; the perimeter is calculated by the number of edge pixels obtained by edge detection.

7. The welding arc detection method based on a pulse camera according to claim 1, characterized in that, Step S7 specifically includes: S71. Calculate the projected radiation intensity. Assuming the welding arc is axisymmetrically distributed, extract the side projection intensity from the arc profile obtained by edge detection. The distribution of , where r is the radial distance from the arc axis to the edge; projected intensity The line integral radiation intensity, obtained by averaging the image intensity along the arc axis, is expressed by the formula: ； in, It is the length of the electric arc axis. This refers to the time index. Below, along the axis coordinates The image intensity value at the radial coordinate r perpendicular to the axis; S72. Calculate radial emissivity using the inverse Abel transform, and then use the inverse Abel transform to calculate the radial emissivity from the projected intensity. Restore radial emissivity It represents the volume emissivity of the electric arc at a radial position r, expressed by the formula: ； in, It is the integral variable; the upper limit of integration ∞ is replaced by the arc radius R in actual calculations, and the radial emissivity at each discrete point is obtained by discretization using numerical integration methods. : ； in, p represents the discrete point index; is the radial step size, and N is the number of discrete points; S73. Calculate the radial temperature distribution: Based on the Stefan-Boltzmann law, the radial emissivity... Convert to radial temperature Assuming the electric arc is a blackbody or a graybody, then The calculation formula is: ； in, It is the Stefan-Boltzmann constant, and the temperature unit is Kelvin.

8. The welding arc detection method based on a pulse camera according to claim 1, characterized in that, Step S8 specifically includes: S81. Using the area and perimeter of consecutive frame images, calculate the temporal change rate of geometric features to quantify the stability of the arc shape; the temporal change rate of geometric features includes the area change rate. and the rate of change of perimeter , defined as the relative difference between the current frame and the previous frame, is expressed by the formula: ； in, It is the arc area in frame t, in pixels². It is the arc area of ​​the previous frame; It is the arc circumference in frame t, in pixels; It represents the arc circumference of the previous frame; |·| represents the absolute value; S82. Calculate the temperature field uniformity; based on radial temperature. Calculate the standard deviation of temperature and temperature gradient This is used to quantify the stability of heat distribution; the formula is expressed as: ； ； Where N is the number of radial discrete points; It is the first p Radial temperature at discrete points; It is the average radial temperature. ; Radial step size; S83, Quantitative Stability Index; Calculate the overall stability index by combining geometric and temperature characteristics. To assess arc stability; the formula is expressed as: ； in, It is the average area change rate across multiple frames; It is the average perimeter change rate across multiple frames; , , It is a weighting coefficient; it sets the arc stability threshold to determine whether welding parameters need to be adjusted.

Images