Reinforcing mesh multi-welding-spot quality detection method and system based on infrared thermal imaging

Abstract

The invention provides a reinforcing mesh multi-welding-spot quality detection method and system based on infrared thermal imaging. Wherein for each crossed steel bar welding spot, the fusion form of the crossed steel bar welding spot is identified based on the thermal radiation image; based on the fusion forms of all the crossed steel bar welding spots, quality detection results of the multiple welding spots of the steel bar mesh are generated, and the quality detection results comprise geometric judgment results used for representing whether the fusion forms meet preset form requirements or not; and according to the quality detection result, a corresponding early warning mode is adopted for self-adaptive early warning, and welding parameters adopted by the next electric resistance welding process are adjusted according to early warning information. According to the technical scheme provided by the invention, millisecond-level multi-welding-spot synchronous quality detection and process closed-loop optimization are realized, and the production yield and the process adaptability of the large reinforcing mesh are greatly improved.

Description

Technical Field

[0001] The present application relates to the technical field of weld quality detection technology, and in particular to a method and system for detecting the quality of multiple welds of a steel mesh based on infrared thermal imaging. Background Art

[0002] In automated resistance welding lines for large steel mesh, the welding cycle is extremely short, requiring full-width weld quality inspection between the end of welding and the start of mesh stretching. This inspection method must complete dozens of welds within a preset timeframe, and the results must be fed back to the welding control system in real time to enable adaptive adjustment of welding parameters.

[0003] The existing solution uses a mobile infrared probe array scanning system. A transversely movable infrared sensor array is deployed above the weld mesh. A robotic arm drives the probes to scan the welds in sections, and the system uses a temperature decay model to predict weld nugget quality. After the mesh is pulled, the system performs spot checks and retests any abnormal areas for verification.

[0004] However, the existing solution has three technical bottlenecks. First, due to timing mismatch, the detection time far exceeds the effective time window. During acquisition, the solder joint temperature has dropped to a low temperature stage, and the key thermal characteristics are lost. Second, the spatial coverage is limited by the scanning path, and only local solder joint sampling detection can be achieved, and the full-width synchronous temperature field cannot be obtained. Third, the dynamic error is amplified due to the vibration of the probe movement, causing the temperature gradient curve to be distorted, and the predicted results of the solder joint shear force are seriously deviated from the true value. Summary of the Invention

[0005] The present application provides a method and system for detecting the quality of multiple weld points of steel mesh based on infrared thermal imaging, which is used to solve the problems in the prior art of low accuracy of multi-weld point quality detection caused by lack of timeliness of thermal features, insufficient full-area coverage capability, and measurement distortion under dynamic interference.

[0006] In a first aspect, the present application provides a method for detecting the quality of multiple weld points of a steel mesh based on infrared thermal imaging, comprising: During the time interval between the end of the resistance welding process and the start of the self-tensioning mesh operation, the infrared thermal imager array synchronously collects thermal radiation images of the surface temperature of all cross-rebar welds, where the cross-rebar welds are welds obtained by the resistance welding process in the steel mesh.

[0007] For each cross steel bar weld point, identifying the fusion morphology of the cross steel bar weld point based on the thermal radiation image;

[0008] Generating a quality inspection result of multiple weld points of the steel mesh based on the fusion morphology of all cross-rebar weld points, wherein the quality inspection result includes a geometric judgment result for indicating whether the fusion morphology meets a preset morphology requirement;

[0009] According to the quality inspection result, a corresponding warning method is used to perform adaptive warning, so as to adjust the welding parameters used in the next resistance welding process according to the warning information.

[0010] Optionally, the fusion morphology includes: pixel area value and nugget diameter;

[0011] The identifying the fusion morphology of the cross steel bar welds based on the thermal radiation image includes: Preprocessing the thermal radiation image to obtain a processed thermal radiation image;

[0012] Identifying a boundary area of ​​a cross-rebar weld from the processed thermal radiation image, and calculating a pixel area value of the boundary area;

[0013] According to the pixel area value of the boundary area, a diameter inversion operation is performed using a pre-established diameter calibration model to obtain the nugget diameter of the cross steel bar weld. The parameters of the diameter calibration model are calibrated according to the steel bar diameter and welding parameters.

[0014] Optionally, identifying the boundary area of ​​the cross steel bar welds from the processed thermal radiation image includes: Identifying, by temperature threshold segmentation, all continuous pixel regions in the processed thermal radiation image whose temperature values ​​are greater than a preset temperature threshold;

[0015] Establishing a topological relationship graph representing pixel connectivity based on the continuous pixel region, converting a boundary node set in the topological relationship graph into a binary image matrix through coordinate mapping, and generating an initial boundary mask;

[0016] Performing hole filling processing on the initial boundary mask to obtain a filled boundary mask;

[0017] performing edge smoothing on the filled boundary mask to obtain a smoothed boundary mask;

[0018] Based on the smoothed boundary mask, the temperature gradient values ​​of the pixels corresponding to the boundary nodes are calculated, and the pixels with temperature gradient values ​​greater than or equal to the preset temperature gradient threshold are screened to determine the boundary area of ​​the cross-rebar weld.

[0019] Optionally, the step of calculating the temperature gradient values ​​of pixels corresponding to the boundary nodes based on the smoothed boundary mask, and screening pixels having temperature gradient values ​​greater than or equal to a preset temperature gradient threshold to determine the boundary area of ​​the cross-rebar weld includes:

[0020] Based on the smoothed boundary mask, the surface temperature of the boundary nodes in the thermal radiation image is obtained. Combined with the surface temperature of the cross-rebar welds in another thermal radiation image, the temperature gradient value of the corresponding pixel of the boundary node is calculated pixel by pixel along the boundary normal direction, where the boundary normal direction is determined by the coordinates of the adjacent nodes in the topological relationship graph;

[0021] Obtain the temperature gradient threshold corresponding to the steel bar diameter from the preset configuration table;

[0022] Pixels with temperature gradient values ​​greater than or equal to the preset temperature gradient threshold are screened to determine the boundary area of ​​the cross-rebar welds.

[0023] Optionally, the diameter inversion operation is performed based on the pixel area value of the boundary region using a pre-established diameter calibration model to obtain the nugget diameter of the cross steel bar weld, and the parameters of the diameter calibration model are obtained by calibration based on the steel bar diameter and welding parameters, including:

[0024] According to the steel bar diameter and welding parameters, the parameters of the diameter calibration model are matched from the preset calibration library;

[0025] According to the pixel area value of the boundary area, combined with the welding position compensation factor, a diameter inversion operation is performed using a pre-established diameter calibration model to obtain the weld core diameter of the cross steel bar weld. The welding position compensation factor is dynamically calculated through the electrode offset.

[0026] Optionally, preprocessing the thermal radiation image to obtain a processed thermal radiation image includes: Separating thermal signals of different frequency bands in the thermal radiation image to remove environmental interference, extract information of the weld nugget edge, and obtain a filtered thermal radiation image;

[0027] Based on the background temperature value of the non-welding area around the weld point in the filtered thermal radiation image, a sensitivity correction value is calculated to eliminate the radiation error caused by the oxidation of the steel bar surface, and a thermal radiation image with the radiation error eliminated is obtained;

[0028] Adaptive histogram equalization is performed on the core area of ​​the solder joint in the thermal radiation image after the radiation error is eliminated to obtain a processed thermal radiation image.

[0029] Optionally, the step of performing adaptive histogram equalization on the core area of ​​the solder joint in the thermal radiation image after the radiation error is eliminated to obtain a processed thermal radiation image includes: Taking the highest temperature point in the thermal radiation image as the center, a circular area with the diameter of the steel bar as the diameter is defined as the core area of ​​the weld;

[0030] In the core area of ​​the solder joint, a preset pixel sliding window is used to traverse all pixels and a probability distribution function of the temperature value in the window is calculated;

[0031] Establishing a mapping relationship from temperature value to equalized grayscale value according to the probability distribution function;

[0032] Transforming the temperature values ​​of all pixels in the window using the mapping relationship;

[0033] Bilinear interpolation is used to fuse the transformation results of adjacent windows, and dynamic range compression processing is performed on the transformed temperature value of the core area of ​​the solder joint to suppress the over-enhancement area, thereby obtaining a processed thermal radiation image.

[0034] In a second aspect, the present application provides a steel mesh multi-weld point quality detection system based on infrared thermal imaging, comprising: An acquisition module is used to synchronously acquire, through an infrared thermal imager array, thermal radiation images of the surface temperatures of all cross-rebar welds obtained by the resistance welding process in the steel mesh during the time interval between the end of the resistance welding process and the start of the self-tensioning operation. The cross-rebar welds are welds obtained by the resistance welding process in the steel mesh.

[0035] an identification module for identifying, for each cross-rebar weld point, a fusion morphology of the cross-rebar weld point based on the thermal radiation image;

[0036] A generation module, configured to generate a quality inspection result of multiple weld points of the steel mesh based on the fusion morphology of all cross-rebar weld points, wherein the quality inspection result includes a geometric judgment result for indicating whether the fusion morphology meets a preset morphology requirement;

[0037] The early warning module is used to perform adaptive early warning using a corresponding early warning method according to the quality detection result, so as to adjust the welding parameters used in the next resistance welding process according to the early warning information.

[0038] In a third aspect, the present application provides a computing device comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a method for quality detection of multiple weld points of steel mesh based on infrared thermal imaging as described in any one of the first aspects.

[0039] In a fourth aspect, the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, it implements a method for detecting multiple weld points of a steel mesh based on infrared thermal imaging as described in any one of the first aspects.

[0040] In the present application, a method for quality inspection of multiple weld points of a steel mesh based on infrared thermal imaging is provided, the method comprising: synchronously collecting, through an infrared thermal imager array, thermal radiation images of the surface temperatures of all cross-rebar weld points within a time interval between the end of a current resistance welding process and the start of a self-pulling mesh action, the cross-rebar weld points being weld points obtained by the current resistance welding process in the steel mesh; for each cross-rebar weld point, identifying the fusion morphology of the cross-rebar weld point based on the thermal radiation image; generating quality inspection results of multiple weld points of the steel mesh based on the fusion morphology of all cross-rebar weld points, the quality inspection results comprising a geometric judgment result for characterizing whether the fusion morphology meets a preset morphology requirement; and performing adaptive warning using a corresponding warning method based on the quality inspection results, so as to adjust the welding parameters used in the next resistance welding process according to the warning information.

[0041] In this application, the problem of thermal feature timeliness is solved by synchronously collecting full-area weld thermal radiation images within a very short time window from the end of resistance welding to the start of mesh pulling, ensuring the capture of transient high-temperature fields; the fusion morphology of each weld is identified based on thermal images, and millisecond-level parallel quantitative analysis of dozens of welds is achieved; global detection results are generated by integrating all weld quality data, and the welding control system is linked to perform adaptive early warning and parameter adjustment, forming a closed-loop control from detection to process optimization, thereby improving the production yield and process stability of large meshes.

[0042] Furthermore, when identifying the fusion morphology of welds based on thermal radiation images, the image is first preprocessed, and then continuous pixel areas with a temperature threshold greater than a preset threshold are extracted through temperature threshold segmentation. A topological relationship diagram of the area is constructed and converted into a binary mask, and an optimized boundary mask is generated through void filling and edge smoothing. Finally, the precise boundary area of ​​the weld is determined based on temperature gradient threshold screening, its pixel area value is calculated, and the weld core diameter value is output using a pre-calibrated diameter inversion model. This solution improves the boundary anti-interference ability through temperature gradient threshold screening and mask optimization processing, effectively suppressing the weld core area identification deviation caused by thermal diffusion. The pre-calibrated model integrates the key variables of steel bar diameter and welding parameters to achieve universal diameter inversion for multi-specification mesh production lines. The full-process algorithm has linear computational efficiency, and the single weld point analysis time is extremely low, meeting the millisecond-level real-time requirements of synchronous detection of high-density weld point arrays.

[0043] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0045] Figure 1 A flowchart of a method for detecting multiple weld points of a steel mesh based on infrared thermal imaging provided in an embodiment of the present application;

[0046] Figure 2 A schematic structural diagram of a steel mesh multi-weld point quality inspection system based on infrared thermal imaging provided in an embodiment of the present application;

[0047] Figure 3 A schematic diagram of the structure of a computing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0048] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0049] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 11, 12, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0050] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0051] In order to solve the problems of low accuracy and low efficiency in the synchronous detection of multi-weld quality caused by the lack of timeliness of thermal features, insufficient full-area coverage and measurement distortion under dynamic interference in the prior art, an embodiment of the present application provides a method for quality detection of multiple welds in steel mesh based on infrared thermal imaging. The method adopts the following ideas: by accurately locking the high-fidelity window period of thermal features from the end of the resistance welding process to the start of the self-pulling mesh action, a spatially distributed infrared thermal imager array is used to synchronously capture the transient temperature field of the full-area welds; a multi-level processing chain based on thermal radiation images is used to realize millisecond-level quantitative analysis of the fusion morphology, and a dynamic calibration model with adaptive process parameters is combined to invert the weld core diameter; quality inspection results are generated by judging the compliance of full-width geometric features, and closed-loop process optimization is triggered based on defect distribution, thereby systematically solving the problems of low accuracy and low efficiency in the synchronous detection of multi-weld quality.

[0052] Figure 1 A flow chart of a method for inspecting the quality of multiple weld points of a steel mesh based on infrared thermal imaging is provided in an embodiment of the present application. Figure 1 As shown, the method includes: S11. During the time interval between the end of the resistance welding process and the start of the self-tensioning mesh action, thermal radiation images of the surface temperatures of all cross-rebar welds are synchronously collected through an infrared thermal imager array. The cross-rebar welds are welds obtained by the resistance welding process in the steel mesh.

[0053] Among them, the resistance welding process refers to a metal processing method that uses electric current to pass through the contact surface of steel bars to generate resistance heat for fusion connection, including the three elements of electrode pressure application, welding current conduction and welding power-on time control. The self-pulling mesh action refers to the mechanical operation of the traction mechanism pulling the steel mesh out of the workstation after welding is completed, and its start mark detection window is closed. The time interval refers to the process gap period from the electrode leaving the weld point to the triggering of the self-pulling mesh mechanism, which is used to accommodate the detection operation. The infrared thermal imager array refers to a spatially distributed imaging system composed of multiple high-frame rate thermal imagers, with a field of view covering the full width of the mesh to achieve temperature field capture. Synchronous acquisition can refer to all thermal imagers triggering exposure at the same time based on a unified timing signal to ensure that the temperature field data of dozens of welds are aligned in time and space. The thermal radiation image is a two-dimensional matrix that records the infrared radiation intensity on the surface of the weld point, and the pixel value can be mapped to the temperature distribution.

[0054] Cross-rebar welds are newly welded connections created at the intersection of rebar within a single resistance welding cycle. A steel mesh is a building component composed of longitudinal and transverse rebar arranged in a grid pattern at regular intervals. For example, a steel mesh is 32 meters long and 12 meters wide.

[0055] In an embodiment of the present application, first, within the time interval between the end of the resistance welding process and the start of the self-pulling mesh action, a synchronous acquisition operation is performed by an infrared thermal imager array covering the entire width of the steel mesh to capture the surface temperature thermal radiation images of all cross-rebar welds at one time; wherein the cross-rebar welds can refer to the welds formed on the steel mesh by this resistance welding process, which is a row of the steel mesh and can be dozens of welds, ensuring that the data timeliness corresponds accurately to the welding position.

[0056] S12. For each cross-rebar weld point, identify the fusion morphology of the cross-rebar weld point based on the thermal radiation image.

[0057] Among them, the fusion morphology can be a composite indicator that characterizes the geometry and metallurgical properties of the weld nugget area, including the weld nugget diameter and the heat-affected zone characteristics.

[0058] In the embodiment of the present application, for each cross-rebar weld, image processing technology is first used to extract the thermal characteristics of the weld, and the spatial distribution of the fusion morphology is determined through temperature gradient analysis and boundary recognition algorithm; specifically: the pixel area of ​​the high-temperature area is calculated and the weld core diameter is inverted, and finally the quantitative fusion morphology parameters including geometric dimensions and thermodynamic characteristics are output.

[0059] S13. Based on the fusion morphology of all cross-rebar welds, generate quality inspection results of multiple welds of the steel mesh. The quality inspection results include geometric judgment results for characterizing whether the fusion morphology meets preset morphology requirements.

[0060] The quality inspection results are structured data that outputs the compliance status of the weld, including the nugget diameter threshold determination and shear force prediction. The geometric judgment result is a binary judgment flag based on the ratio between the effective nugget diameter and the rebar diameter.

[0061] In an embodiment of the present application, based on the fusion morphology data of all cross-rebar welds, the geometric judgment results of each weld are first compared with the preset morphology requirements; then, a spatial aggregation algorithm is used to generate a quality inspection result covering the entire steel mesh, which includes a binary judgment of the compliance of the fusion morphology of each weld and a heat map of the overall quality distribution.

[0062] S14. Based on the quality inspection results, a corresponding warning method is used to perform adaptive warning, so as to adjust the welding parameters used in the next resistance welding process according to the warning information.

[0063] Early warning methods can include alarm mechanisms graded by defect severity, including audible and visual warnings, parameter adjustment instructions, and defect location marking. Adaptive early warning is an intelligent feedback mechanism that dynamically selects warning levels based on real-time inspection results and triggers process optimization. Early warning information can be a composite data package containing defect location coordinates, anomaly type, and recommended process correction parameters. Welding parameters are the core variables that control resistance welding quality, including electrode pressure, welding current, and welding power-on time.

[0064] In an embodiment of the present application, based on the quality inspection results, the warning method is first matched according to the defect level and an adaptive warning signal is triggered; then the warning information is fed back to the welding control system in real time, and the welding current, electrode pressure and welding power-on time of the next resistance welding process are dynamically optimized to form a process closed-loop control chain.

[0065] Alternatively, the following is another method for inspecting the quality of multiple weld points of steel mesh based on infrared thermal imaging, which includes the following steps: Step 1: Within the 0.3-0.8 second time window after the resistance welding electrode is separated from the steel mesh and before the mesh pulling action is started, the thermal radiation images of all welds are synchronously collected by an infrared thermal imager array arranged above the weld mesh plane. The field of view of the infrared thermal imager array can cover a weld mesh width of not less than 2 meters, and the imaging frame rate can be not less than 200 Hz to capture the transient temperature field distribution during the cooling process of the weld.

[0066] The steel mesh can be a large mesh with a maximum diameter of 32 meters and a maximum width of 12 meters. The spacing between steel bars can be 0.125 meters, so the number of multiple welds can be 12÷0.125=96. In the process of resistance spot welding, it should include three stages: pre-pressing, power-on, and forging. In the power-on stage, the following welding parameters are used: electrode pressure, welding current, and welding power-on time. Multiple welds are welded simultaneously. After welding, there is a holding time in the forging stage. Large meshes can be selected according to the diameter of the steel bars, with appropriate electrode pressure, welding current, and welding power-on time. During the welding process, the local high temperature is above 1000 degrees, and the steel bars are stacked, so after welding, whether there is de-soldering or welding leakage, and whether the weld is too large or too small, it is not easy to find. For the embodiment of the present application, the synchronous detection of the quality of multiple welds can be achieved based on thermal imaging data.

[0067] Step 2: The detection system communicates with the welding control system in real time through the industrial bus, automatically triggers image acquisition upon receiving the welding timing signal, and stores the thermal imaging data in a temporally and spatially matched manner with the welding position coordinates.

[0068] Step 3: Perform image analysis based on a pre-established weld temperature gradient calibration model. The calibration model establishes the corresponding temperature gradient influence diagram range and weld nugget morphology standard feature library based on the parameter combination of steel bar diameter 12 mm, welding current 25 kA, and power-on time 2.5 s.

[0069] Specifically, the "temperature gradient" is clearly divided into two indicators: the weld core diameter and the shear resistance, which correspond to the spatial distribution and time evolution characteristics of the thermal image respectively. Among them, the process of determining the weld core diameter is: according to the combination of the parameters of the steel bar diameter, welding current and power-on time, a linear regression equation is established between the area of ​​the high-temperature zone of the weld with a temperature ≥ 1000°C and the weld core diameter; for the prediction of shear resistance, the embodiment of the present application can associate the cooling rate of the weld temperature gradient curve with the metal bonding strength, calibrate the temperature drop slope threshold, and determine that the shear resistance is insufficient when the slope is lower than the critical value. This step 3 can be understood as the model building stage, in which a mapping model between temperature and mass is established through experimental calibration to provide a criterion for the real-time detection of step 4.

[0070] Step 4: Use the edge computing unit to calculate the three-dimensional temperature field distribution of each weld in real time, extract the kurtosis coefficient and the descent rate characteristic value of the temperature gradient curve, and combine the critical threshold in the calibration model to determine whether the weld core diameter is smaller than the steel bar diameter.

[0071] Specifically, the nugget diameter calculation process involves extracting the area of ​​pixels with temperatures ≥1000°C in the weld thermal image, calculating the nugget diameter based on a calibration equation, and determining whether it is smaller than the rebar diameter. The shear resistance assessment process involves extracting a temperature gradient curve along the weld radial direction and calculating the slope between 600°C and 300°C during the cooling phase. If the slope is below a calibration threshold, such as 0.15°C / ms, the defect is identified as a shear resistance defect. Step 4 is the model application phase. During production line inspection, the nugget diameter calculation and shear resistance assessment algorithms are executed in real time based on the model parameters established in Step 3 to complete quality assessment.

[0072] Step 5: When an abnormal solder joint is detected, a quality warning signal is issued through the production line PLC system before the next welding cycle begins, and the defect location coordinates and temperature characteristic parameters are automatically recorded. At the same time, the inkjet printer is controlled to physically mark the defective solder joint.

[0073] For example, when any of the indicators, such as the nugget diameter or shear resistance, is abnormal, a graded warning (light, medium, or severe defects) is triggered through the production line PLC system before the next welding cycle. The robotic arm is synchronously controlled to perform laser marking and positioning on the defective welds, and the abnormal data is associated with the welding process parameter library to generate a process optimization recommendation report.

[0074] In summary, the embodiment of the present application first collects the thermal radiation image of the entire steel mesh when the electrode is lifted after the resistance welding process is completed, through an infrared thermal imager array spanning a width of twelve meters, to capture the transient temperature field of all cross-steel welds. Secondly, for each weld thermal image, the temperature gradient threshold segmentation and mask optimization algorithm are used to extract the weld nugget boundary area, calculate the pixel area and invert the weld nugget diameter value. Then, the geometric judgment results of all welds are aggregated to generate a quality inspection report containing the weld nugget diameter compliance and shear resistance prediction. Finally, a graded warning is triggered based on the defect distribution, the current parameters of the next welding cycle are synchronously adjusted, and the location of the problem welds is marked, completing the closed-loop control from detection to process optimization.

[0075] By executing S11 to S14, the embodiment of the present application solves the problem of capturing the rapid attenuation of thermal characteristics of high-temperature welds by accurately locking the detection time window between welding and mesh drawing; uses parallel image processing to achieve millisecond-level quantitative analysis of the fusion morphology of dozens of welds; and generates process optimization instructions based on full-width quality data, effectively improving the welding qualification rate of large-scale steel mesh and the intelligent level of the production line.

[0076] In a possible embodiment, the fusion morphology includes: pixel area value and nugget diameter. S12, for each cross-rebar weld, based on the thermal radiation image, identifying the fusion morphology of the cross-rebar weld, including: Step 121 : pre-process the thermal radiation image to obtain a processed thermal radiation image.

[0077] Among them, preprocessing is a sequence of operations to suppress noise, correct radiation errors and enhance contrast of the original thermal radiation image, providing an optimized data basis for subsequent analysis.

[0078] In an embodiment of the present application, as described in subsequent steps c1 to c3, frequency domain filtering is used to separate thermal signals of different frequency bands in the thermal radiation image and extract the edge information of the weld core; a sensitivity correction value is calculated based on the background temperature of the non-welding area around the weld in the filtered thermal radiation image to eliminate the radiation error caused by surface oxidation of the steel bar; finally, adaptive histogram equalization is performed in the core area of ​​the weld to enhance the temperature distribution contrast, and finally the processed thermal radiation image is output.

[0079] For example, in the filtered thermal radiation image, with the center of the cross steel bar weld as the origin, a circular area with a radius of 3-5 times the diameter of the steel bar is delineated as a candidate area to ensure that the area is not affected by welding heat; continuous pixel blocks with uniform temperature distribution (standard deviation of temperature ≤ 5°C) are screened from the candidate area, and the area of ​​the pixel block is not less than 100 pixels as the non-welding reference area; the average temperature of all pixels in the non-welding reference area is calculated and recorded as the background measurement temperature ; Use the thermocouple data collected synchronously with the infrared thermal imager as the benchmark. Specifically: Before the production of the steel mesh, preset the thermocouple sensor in the non-welding area, and its measurement point coincides with the spatial position of the non-welding reference area in the field of view of the infrared thermal imager; After the resistance welding process is completed, the actual temperature measured by the thermocouple is synchronously recorded and recorded as the actual background temperature ; Standard emissivity of unoxidized steel bars , pre-calibrated by experiments, the oxidation of the steel surface will cause the infrared emissivity ε to deviate from the standard emissivity , which causes the temperature measured by the infrared thermal imager to deviate from the actual temperature. The calculation formula for the sensitivity correction value K1 is: K1= K1 represents the radiation measurement deviation coefficient caused by surface oxidation and can be used as a sensitivity correction value: when K1 = 1, there is no radiation error; when K1 > 1, the infrared thermal imager's measurement value is low; when K1 < 1, the infrared thermal imager's measurement value is high. Multiplying the temperature value of each pixel in the filtered thermal radiation image by the sensitivity correction value K1 produces a thermal radiation image with radiation error eliminated, thereby eliminating the radiation error caused by steel bar surface oxidation.

[0080] Step 122: Identify the boundary area of ​​the cross steel bar welds from the processed thermal radiation image, and calculate the pixel area value of the boundary area.

[0081] The boundary region is the enclosed area around the weld nugget, precisely defined by the temperature gradient threshold. Its spatial extent determines the accuracy of the weld nugget diameter calculation. The pixel area value refers to the total number of valid pixels within the weld nugget boundary region in the thermal image, reflecting the projection coverage of the weld nugget on the two-dimensional plane.

[0082] In an embodiment of the present application, the boundaries of cross-rebar welds are first identified from a preprocessed thermal radiation image: a temperature threshold segmentation technique is used to extract high-temperature continuous pixel areas, a pixel connectivity topology map is constructed, and an initial boundary mask is generated; then, hole filling and edge smoothing operations are performed on the mask; finally, the precise boundary area is determined based on temperature gradient threshold screening, and the pixel area value of the area is calculated.

[0083] Step 123: Based on the pixel area value of the boundary area, a diameter inversion operation is performed using a pre-established diameter calibration model to obtain the nugget diameter of the cross-rebar weld. The parameters of the diameter calibration model are calibrated according to the rebar diameter and welding parameters.

[0084] The diameter calibration model is a mathematical mapping relationship established between rebar specifications and welding process parameters, used to convert pixel area into physical dimensions. The diameter inversion operation can be understood as the process of using the linear regression equation in the calibration model to reversely derive the boundary pixel area value into the actual weld nugget diameter. The weld nugget diameter represents the key geometric dimension of the effective bonding surface of the resistance spot weld fusion zone and must reach a specific ratio to the rebar diameter to meet strength requirements. Welding parameters include welding current, welding power-on time, electrode pressure, etc.

[0085] In an embodiment of the present application, first, the corresponding diameter calibration model parameters are matched from a preset calibration library according to the steel bar diameter, welding current and welding power-on time; then, the pixel area value of the boundary area is combined with the welding position compensation factor and input into the diameter calibration model; finally, the actual physical size value of the weld core diameter is output through inversion operation.

[0086] The following is a specific example: First, the captured thermal radiation image is preprocessed: ambient thermal noise is filtered out through frequency band separation, radiation sensitivity errors are corrected using background temperature values, and adaptive histogram equalization is performed on the core area of ​​the weld. Next, the weld boundary is extracted from the processed image: a temperature threshold is applied to segment and lock the high-temperature continuous region. After mask generation, void filling, and edge smoothing, the final boundary region is determined based on the temperature gradient threshold and its pixel area value is calculated. Finally, the corresponding calibration model is called based on the rebar diameter and welding current parameters. The pixel area value is combined with the electrode offset compensation factor and input into the inversion equation, outputting the actual weld nugget diameter value for quality judgment.

[0087] By executing steps 121 to 123, the embodiment of the present application eliminates thermal radiation interference through refined preprocessing to ensure the reliability of boundary recognition; achieves sub-pixel positioning of the weld nugget area based on temperature gradient threshold and mask optimization technology; and combines the process parameter adaptive calibration model to complete the physical size inversion of the weld nugget diameter in a very short time, providing a core geometric basis for welding quality judgment.

[0088] In a possible embodiment, step 122, identifying the boundary area of ​​the cross-rebar weld from the processed thermal radiation image, includes: Step a1: using a temperature threshold segmentation method, identifying all continuous pixel regions in the processed thermal radiation image whose temperature values ​​are greater than a preset temperature threshold.

[0089] Temperature threshold segmentation is a binary processing method that divides a thermal image into a target high-temperature area and a background area based on a preset temperature threshold. A continuous pixel region is a connected image subset consisting of spatially adjacent pixels that meet the temperature condition.

[0090] In an embodiment of the present application, temperature threshold segmentation technology is first used to process the thermal radiation image, the temperature value of each pixel is compared with a preset temperature threshold, and all pixels exceeding the threshold are screened out; then, connected pixel clusters are identified based on the spatial adjacency relationship of pixels to form a set of continuous pixel areas representing high-temperature areas.

[0091] Step a2: Based on the continuous pixel area, a topological relationship graph representing pixel connectivity is established, and the boundary node set in the topological relationship graph is converted into a binary image matrix through coordinate mapping to generate an initial boundary mask.

[0092] A topological graph is a graph data structure that describes pixel adjacency relationships, with nodes representing pixel locations and edges representing spatial connectivity. Coordinate mapping is a spatial transformation that converts node coordinates in a topological graph into indices of a two-dimensional image matrix. A boundary node set is a set of pixel coordinates located at the edge of a region's outline in a topological graph. A binary image matrix is ​​a digital matrix containing only two values: 0 and 1, where 0 represents background and 1 represents target. An initial boundary mask is an initial binary outline template generated from a boundary node set.

[0093] In an embodiment of the present application, a topological relationship diagram is first established for a continuous pixel area, and the set of region boundary nodes is determined by analyzing the adjacency between pixels; secondly, the boundary nodes in the topological diagram are converted into a two-dimensional coordinate matrix using coordinate mapping technology; finally, a binary image matrix consisting of 0 and 1 is generated, where the 1-value area corresponds to the initial boundary mask.

[0094] Step a3: Perform hole filling processing on the initial boundary mask to obtain a filled boundary mask.

[0095] Hole filling is a morphological operation that fills the closed blank areas within the binary mask with pixel values. The filled boundary mask is the complete area outline template after eliminating the internal holes.

[0096] In an embodiment of the present application, the closed blank areas in the initial boundary mask are first scanned; secondly, a pixel filling operation is performed on the unconnected areas inside the mask; and finally, a filled boundary mask with internal holes eliminated is output.

[0097] Step a4: performing edge smoothing on the filled boundary mask to obtain a smoothed boundary mask.

[0098] Edge smoothing is an image optimization process that uses filtering algorithms to eliminate jagged edges. The smoothed edge mask is an optimized contour template with continuous and smooth edge features.

[0099] In the embodiment of the present application, the jagged edges of the filled boundary mask are first detected; secondly, a Gaussian convolution kernel is used to perform smoothing filtering along the boundary; and finally, a natural boundary contour with continuous edge transition is generated, i.e., the boundary mask after smoothing.

[0100] Step a5: Based on the smoothed boundary mask, calculate the temperature gradient values ​​of the pixels corresponding to the boundary nodes, and screen pixels whose temperature gradient values ​​are greater than or equal to a preset temperature gradient threshold to determine the boundary area of ​​the cross-rebar weld.

[0101] The temperature gradient represents the rate of temperature change per unit spatial distance and is calculated by the temperature difference between adjacent pixels. The temperature gradient threshold, which determines the critical value of the heat conduction characteristic at the nugget boundary, is set based on the material's thermophysical properties.

[0102] In an embodiment of the present application, first, the boundary node pixels are located based on the smoothed boundary mask; secondly, the temperature change rate of adjacent pixels in the node normal direction is calculated as the temperature gradient value; finally, the pixels whose temperature gradient values ​​exceed the preset temperature gradient threshold are screened to determine the final boundary area of ​​the solder joint.

[0103] The following is a specific example: First, temperature threshold segmentation is used to extract high-temperature continuous pixel regions from a thermal image. Next, a topological relationship graph of the pixels in this region is established, and the boundary node set is mapped into a binary matrix to generate an initial boundary mask. A hole-filling operation is then performed on this mask to eliminate internal gaps, and edge smoothing is performed to obtain a continuous contour. Finally, the temperature gradient values ​​of the boundary nodes are calculated based on the smoothed mask, and nodes exceeding the threshold are screened to determine the precise boundary region of the solder joint.

[0104] By executing steps a1 to a5, the embodiment of the present application effectively suppresses the boundary blurring phenomenon caused by thermal diffusion through a multi-level mask optimization and temperature gradient screening mechanism, achieves sub-pixel precise positioning of the weld core edge, and provides a high-precision spatial reference for subsequent geometric parameter calculations.

[0105] In a possible embodiment, step a5, calculating the temperature gradient values ​​of pixels corresponding to the boundary nodes based on the smoothed boundary mask, and screening pixels having temperature gradient values ​​greater than or equal to a preset temperature gradient threshold to determine the boundary area of ​​the cross-rebar weld, includes:

[0106] Step a51: Based on the smoothed boundary mask, obtain the surface temperature of the boundary node in the thermal radiation image, and combine it with the surface temperature of the cross steel bar weld in another thermal radiation image to calculate the temperature gradient value of the pixel corresponding to the boundary node pixel by pixel along the boundary normal direction, where the boundary normal direction is determined according to the coordinates of the adjacent nodes in the topological relationship diagram.

[0107] Surface temperature refers to the surface radiation temperature of the solder joint, as captured by the infrared thermal imager, reflecting the thermal energy distribution within the weld area. The boundary normal direction is a spatial vector perpendicular to the boundary contour, calculated by rotating the tangent line 90 degrees using the coordinates of adjacent nodes in the topology graph. Adjacent nodes are adjacent pixels directly connected to the target node in the topology graph and are used to determine local geometric features.

[0108] In an embodiment of the present application, first, the boundary node positions are located based on the smoothed boundary mask, and the surface temperature values ​​of these nodes in the current thermal radiation image are obtained; secondly, combined with the surface temperature data of the same solder joint in the thermal radiation image collected at another moment, the temperature change is calculated pixel by pixel along the boundary normal direction; the boundary normal direction is determined by the geometric relationship of the coordinates of adjacent nodes in the topological relationship diagram, and finally the temperature gradient value of each boundary node is output.

[0109] Step a52: Obtain the temperature gradient threshold corresponding to the steel bar diameter from the preset configuration table.

[0110] Among them, the preset configuration table refers to a structured database that stores the mapping relationship between steel bar diameter and temperature gradient threshold, which is pre-calibrated and established based on the thermal conductivity characteristics of the material.

[0111] In an embodiment of the present application, first, the preset gradient threshold configuration table is queried according to the steel bar diameter specifications corresponding to the cross steel bar welds; secondly, the temperature gradient threshold parameters that strictly match the steel bar diameter are extracted from the configuration table to provide a critical benchmark for subsequent boundary judgment.

[0112] Step a53: Filter pixels whose temperature gradient values ​​are greater than or equal to a preset temperature gradient threshold value to determine the boundary area of ​​the cross steel bar welds.

[0113] In an embodiment of the present application, the temperature gradient value calculated at the boundary node is first compared with the obtained temperature gradient threshold; secondly, all valid pixels whose temperature gradient values ​​are greater than or equal to the threshold are screened; finally, these pixel coordinate sets are determined as the final precise boundary area of ​​the cross-rebar weld.

[0114] The following is a specific example: First, the surface temperature of the boundary node in the current frame is obtained based on the smoothed boundary mask. The gradient value is then calculated pixel by pixel along the boundary normal, combined with the temperature data at the same location in the previous frame. Next, the gradient threshold corresponding to the rebar diameter is obtained from a preset configuration table. Finally, node pixels with gradient values ​​exceeding the threshold are screened to determine the precise boundary area of ​​the weld for subsequent quality analysis.

[0115] By executing steps a51 to a53, the embodiment of the present application effectively distinguishes the physical boundary between the weld nugget zone and the heat-affected zone by dynamically calculating the boundary normal temperature gradient; combined with the threshold matching mechanism adaptive to the steel bar specifications, the physical accuracy of boundary positioning is improved, laying a spatial benchmark for the calculation of the weld nugget diameter.

[0116] In one possible embodiment, step 123, based on the pixel area value of the boundary region, a diameter inversion operation is performed using a pre-established diameter calibration model to obtain the nugget diameter of the cross-rebar weld. The parameters of the diameter calibration model are calibrated according to the rebar diameter and welding parameters, including: Step b1: According to the steel bar diameter and welding parameters, the parameters of the diameter calibration model are matched from the preset calibration library.

[0117] Among them, the preset calibration library refers to a database that stores calibration parameters under different combinations of steel bar specifications and welding processes, including the mapping relationship between weld core diameter and pixel area and the error correction coefficient.

[0118] In an embodiment of the present application, first, the values ​​of the steel bar diameter associated with the cross steel bar weld, the intensity values ​​of the welding current, and the welding power-on time are extracted; secondly, the three types of parameters are combined into an index key; then, a fully matching diameter calibration model parameter record is retrieved in the preset calibration library; finally, the regression equation coefficient term and constant term parameter set in the record are obtained.

[0119] Step b2: Based on the pixel area value of the boundary area and the welding position compensation factor, a diameter inversion operation is performed using a pre-established diameter calibration model to obtain the nugget diameter of the cross-rebar weld. The welding position compensation factor is dynamically calculated through the electrode offset.

[0120] Among them, the welding position compensation factor is used to correct the proportional coefficient of the nugget deformation error caused by the electrode position deviation. Divide by the base distance reference value The electrode offset represents the spatial distance deviation between the actual position of the welding electrode and the theoretical center point, and is dynamically measured by the displacement sensor. The example expression of the pre-established diameter calibration model is: ,in, is the physical size of the nugget diameter, is the pixel area value of the boundary area, is the welding position compensation factor, is the proportionality coefficient, is the corrected intercept.

[0121] In an embodiment of the present application, first, the planar distance between the center point of the electrode and the theoretical welding position is measured in real time by an electrode displacement sensor as the electrode offset; secondly, the electrode offset value is divided by a preset base distance reference value to obtain a welding position compensation factor; then, the pixel area value of the boundary area is multiplied by the compensation factor to obtain the effective pixel area after geometric correction; finally, the effective pixel area is input into a pre-established diameter calibration model to perform an inversion calculation, and the actual physical size value of the weld core diameter is output.

[0122] The following is a specific example: First, the corresponding model parameters are matched in the calibration library based on the rebar diameter and welding current parameters. Next, the electrode offset is collected in real time to calculate the position compensation factor. The boundary pixel area value is multiplied by this factor to obtain the corrected area. Finally, the calibration model is used to invert the calculation to output the weld nugget diameter value.

[0123] By executing steps b1 to b2, the embodiment of the present application solves the model generalization problem of multi-specification mesh production through a calibration parameter matching mechanism with process parameter adaptation; introduces dynamic position compensation to correct the spatial distortion of thermal images and improve the accuracy of weld core diameter inversion.

[0124] In a possible embodiment, step 121, preprocessing the thermal radiation image to obtain a processed thermal radiation image, includes: Step c1: separating the thermal signals of different frequency bands in the thermal radiation image to remove environmental interference, extract the edge information of the weld nugget, and obtain a filtered thermal radiation image.

[0125] The thermal signal refers to the infrared radiation energy data in a thermal radiation image that characterizes the surface temperature distribution of an object. Environmental interference refers to non-actual temperature signals caused by non-target heat sources. Nugget edge information refers to the sudden change in thermal distribution at the transition boundary between the molten zone of the solder joint and the base material. Filtering refers to data processing operations that suppress noise and enhance the target signal through frequency or spatial domain transformations.

[0126] In an embodiment of the present application, a frequency domain wavelet transform is first performed on the thermal radiation image to separate the thermal signal components of different frequency bands; secondly, a band-stop filter is designed to suppress the characteristic frequency band corresponding to the environmental interference; then, the high-frequency component containing the weld core edge information is extracted for signal enhancement; finally, the image is reconstructed by inverse transformation to obtain a filtered thermal radiation image with interference removed and edge features enhanced.

[0127] Step c2: Calculate a sensitivity correction value based on the background temperature value of the non-welding area around the weld in the filtered thermal radiation image to eliminate the radiation error caused by steel bar surface oxidation, and obtain a thermal radiation image after the radiation error is eliminated.

[0128] The non-weld area refers to the reference area on the rebar surface around the weld that is unaffected by heat. The background temperature value refers to the statistical value of pixel temperatures in the non-weld area and is used as a reference for radiation error correction. The sensitivity correction value is a temperature calibration factor calculated based on differences in the radiation characteristics of the material surface. Radiation error refers to the temperature measurement deviation caused by the infrared emissivity deviating from the theoretical value due to oxidation of the material surface.

[0129] In an embodiment of the present application, first, the non-welding area outside the solder joint is identified in the filtered image; secondly, the average temperature of all pixels in the area is calculated as the background temperature value; then, the background temperature value is compared with the standard blackbody radiation temperature, and the ratio of the two is calculated as the sensitivity correction value; finally, the original temperature value of each pixel in the entire image is divided by the correction value to eliminate the radiation error caused by surface oxidation, thereby generating a thermal radiation image after the radiation error is eliminated.

[0130] Step c3: performing adaptive histogram equalization on the core area of ​​the solder joint in the thermal radiation image after the radiation error is eliminated, so as to obtain a processed thermal radiation image.

[0131] Among them, histogram equalization refers to a nonlinear transformation method that enhances image contrast by redistributing pixel grayscale values. The core area of ​​the weld spot refers to the projection range of the weld nugget and the adjacent heat-affected zone formed during resistance spot welding on the thermal radiation image. In the thermal radiation image after radiation errors are eliminated, the coordinates of the pixel with the highest temperature are located through global scanning as the center of the weld nugget. This point maintains high temperature characteristics for a very short time after welding due to the slowest heat dissipation rate in the weld nugget. Then, a circular analysis area is defined with this center point as the center and a radius of half the current rebar diameter. The circular analysis area covers the entire physical range of the weld nugget and heat-affected zone, forming the weld spot core area.

[0132] In an embodiment of the present application, first, in the image after the radiation error is eliminated, the core area of ​​the circular weld is delineated with the pixel point with the highest temperature as the center and the steel bar diameter as the diameter; secondly, a pixel sliding window of a preset size is used to traverse the core area, and the probability distribution function of the temperature value in each window is calculated; then a piecewise linear mapping relationship from the temperature value to the equalized grayscale value is established; then all pixel values ​​in the window are converted according to the mapping table; finally, bilinear interpolation fusion is performed on the overlapping window area, and logarithmic dynamic range compression is performed on the conversion result to output the processed thermal radiation image.

[0133] The following is a specific example: First, the original thermal image is band-separated to extract the weld nugget edge features and filter out ambient noise. Next, the background temperature average is calculated for the non-welded area surrounding the solder joint in the filtered image. This is used to derive a sensitivity correction to eliminate radiation errors. Finally, sliding window histogram equalization is performed on the core area of ​​the solder joint, and dynamic range compression is used to output an optimized thermal image.

[0134] By executing steps c1 to c3, the embodiment of the present application effectively separates the target thermal signal from the environmental noise through a multi-level preprocessing mechanism; accurately compensates for the radiation error based on the background temperature of the non-welding area; and combines adaptive contrast enhancement technology to provide a high signal-to-noise ratio thermal image foundation for weld nugget boundary identification.

[0135] In a possible embodiment, step c3, performing adaptive histogram equalization on the core area of ​​the solder joint in the thermal radiation image after the radiation error is eliminated to obtain a processed thermal radiation image, includes:

[0136] Step c31: With the highest temperature point in the thermal radiation image as the center, a circular area with the diameter of the steel bar as the diameter is defined as the core area of ​​the weld point.

[0137] The highest temperature point is the pixel location with the highest radiation intensity in the thermal radiation image, reflecting the center of the weld pool. A circular area, a geometric range with a specified point as the center and a fixed length as the diameter, is used to define the core analysis area. The weld core area is the critical analysis range encompassing the weld nugget and the heat-affected zone, and its diameter is correlated to the rebar size.

[0138] In an embodiment of the present application, all pixels of the thermal radiation image are first traversed, and the single pixel point with the largest temperature value is identified as the highest temperature point; secondly, the coordinates of this point are used as the center of the circle, and the physical size value of the current steel bar diameter is used as the diameter, and a circular analysis area is delineated in the image space; finally, the circular area is defined as the core area of ​​the weld.

[0139] Step c32: In the core area of ​​the solder joint, a preset pixel sliding window is used to traverse all pixels and calculate the probability distribution function of the temperature value in the window.

[0140] The pixel sliding window is a rectangular analysis unit that moves across the image and is used to perform local statistical operations. The probability distribution function is a statistical model that describes the frequency of temperature values ​​within the window and is the basis for histogram equalization.

[0141] In an embodiment of the present application, a rectangular pixel sliding window with fixed length and width is first set; secondly, starting from the upper left corner of the core area of ​​the solder joint, the window is moved in the order of row first and column later until all pixels are covered; then, at each window position, a distribution histogram of the temperature values ​​of all pixels in the window is statistically calculated; finally, a distribution function describing the temperature probability density is generated based on the histogram.

[0142] Step c33: Establish a mapping relationship from the temperature value to the equalized grayscale value according to the probability distribution function.

[0143] The mapping relationship refers to a conversion rule table between the original temperature value and the equalized grayscale value, which is established based on the probability distribution characteristics.

[0144] In an embodiment of the present application, the cumulative distribution value of the probability distribution function is first calculated; secondly, the cumulative distribution value is normalized to the target grayscale range; and then a piecewise linear mapping function table between the original temperature input value and the equalized grayscale output value is constructed.

[0145] Step c34: transform the temperature values ​​of all pixels in the window using the mapping relationship.

[0146] In an embodiment of the present application, the original temperature values ​​of all pixels at the current coverage position of the sliding window are first read; secondly, the target grayscale value corresponding to each temperature value is queried in the mapping function table; finally, the original temperature data is replaced with the mapped grayscale value data.

[0147] Step c35: bilinear interpolation is used to fuse the transformation results of adjacent windows, and dynamic range compression processing is performed on the transformed temperature values ​​of the core area of ​​the solder joint to suppress over-enhancement areas, thereby obtaining a processed thermal radiation image.

[0148] Bilinear interpolation is a smoothing algorithm that uses the weighted values ​​of four adjacent pixels to calculate the new pixel values ​​in the overlapping area. Dynamic range compression uses nonlinear transformations to suppress the contrast enhancement in high-grayscale areas. Overenhancement refers to distorted areas of the image where the local contrast is abnormally amplified due to histogram equalization.

[0149] In an embodiment of the present application, first, for the pixels in the overlapping area of ​​adjacent sliding windows, bilinear weight calculation is performed according to their distance to the centers of four adjacent windows, and the transformation results of different windows are fused; secondly, a logarithmic function transformation is performed on the full-width grayscale image after the core area transformation to compress the dynamic range of the high grayscale value area; finally, a processed thermal radiation image with suppressed over-enhancement phenomenon is output.

[0150] The following is a specific example: First, the highest temperature point in the thermal image is located and a circular core area is delineated, sized to the rebar diameter. Next, a sliding window is used to traverse the core area, calculating the temperature probability distribution within each window and establishing a grayscale mapping relationship. Pixel values ​​within the window are then converted according to the mapping table. Bilinear interpolation is then performed on overlapping areas. Finally, dynamic range compression is applied to output an optimized thermal radiation image for boundary recognition.

[0151] By executing steps c1 to c3, the embodiment of the present application enhances the recognizability of micro-temperature difference features in the weld nugget area through adaptive equalization; uses bilinear interpolation to eliminate window boundary mutations; combines dynamic range compression to suppress over-enhancement distortion, and provides high-fidelity temperature distribution data for weld nugget boundary identification.

[0152] Figure 2This is a structural diagram of a steel mesh multi-weld point quality detection system based on infrared thermal imaging provided in an embodiment of the present application, as shown in FIG. Figure 2 As shown, the system includes:

[0153] The acquisition module 21 is used to synchronously acquire thermal radiation images of the surface temperature of all cross-rebar welds through an infrared thermal imager array during the time interval between the end of the resistance welding process and the start of the self-pulling mesh action. The cross-rebar welds are welds obtained by the resistance welding process in the steel mesh.

[0154] The identification module 22 is configured to identify the fusion morphology of each cross-rebar weld point based on the thermal radiation image.

[0155] The generation module 23 is used to generate quality inspection results of multiple weld points of the steel mesh based on the fusion shape of all cross steel bar weld points. The quality inspection results include geometric judgment results used to characterize whether the fusion shape meets the preset shape requirements.

[0156] The early warning module 24 is used to perform adaptive early warning using a corresponding early warning method according to the quality inspection result, so as to adjust the welding parameters used in the next resistance welding process according to the early warning information.

[0157] Figure 2 The steel mesh multi-weld point quality detection system based on infrared thermal imaging can be performed Figure 1 The implementation principle and technical effects of the infrared thermal imaging-based method for inspecting the quality of multiple weld points in steel mesh are not described in detail here. The specific manner in which the various modules and units in the infrared thermal imaging-based system for inspecting the quality of multiple weld points in steel mesh have been described in detail in the embodiments of the method and will not be elaborated on here.

[0158] In one possible design, Figure 2 The embodiment shown is a steel mesh multi-weld point quality detection system based on infrared thermal imaging that can be implemented as a computing device, such as Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32 .

[0159] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32 .

[0160] The processing component 32 is used to: control the flight altitude and heading angle of the detection drone, so that the laser radar and thermal infrared sensor deployed on the detection drone respectively collect three-dimensional point cloud data of the vertical section of the wildfire area and the thermal radiation distribution data of the wildfire area at multiple angles; combine the three-dimensional point cloud data with the thermal radiation distribution data to generate a target fire spread path; use the embedded real-time processing chip deployed in the detection drone to process the thermal radiation distribution data in real time, identify temperature abnormality areas and remove interference areas, and generate fire point locations and classification results; based on the target fire spread path, the fire point location and the classification results, send dynamic scheduling instructions to the fire-fighting drone cluster through a low-latency communication link to adjust the location, fire extinguishing dosage and coverage range of the fire-fighting drones in the fire-fighting drone cluster.

[0161] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above method. Of course, the processing component may also be implemented as 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 perform the above method.

[0162] The storage component 31 is configured to store various types of data to support operations on the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as random access memory (RAM), 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.

[0163] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0164] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0165] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0166] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0167] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is a method for detecting the quality of multiple weld points of a steel mesh based on infrared thermal imaging.

[0168] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0169] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0170] Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented using software plus a necessary general-purpose hardware platform, or of course, hardware. Based on this understanding, the essence of the above technical solution, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, or an optical disk, and includes a number of instructions for causing a computer device (such as a personal computer, server, or network device) to execute the methods described in each embodiment or certain portions of the embodiments.

[0171] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A method for detecting the quality of multiple weld points of steel mesh based on infrared thermal imaging, characterized in that: include: During the time interval between the end of the resistance welding process and the start of the self-tensioning mesh operation, the infrared thermal imager array synchronously collects thermal radiation images of the surface temperature of all cross-rebar welds, where the cross-rebar welds are welds obtained by the resistance welding process in the steel mesh. For each cross steel bar weld point, identifying the fusion morphology of the cross steel bar weld point based on the thermal radiation image; Generating a quality inspection result of multiple weld points of the steel mesh based on the fusion morphology of all cross-rebar weld points, wherein the quality inspection result includes a geometric judgment result for indicating whether the fusion morphology meets a preset morphology requirement; According to the quality inspection result, a corresponding warning method is used to perform adaptive warning, so as to adjust the welding parameters used in the next resistance welding process according to the warning information.

2. The method according to claim 1, characterized in that The fusion morphology includes: pixel area value and nugget diameter; The identifying the fusion morphology of the cross steel bar welds based on the thermal radiation image includes: Preprocessing the thermal radiation image to obtain a processed thermal radiation image; Identifying a boundary area of ​​a cross-rebar weld from the processed thermal radiation image, and calculating a pixel area value of the boundary area; According to the pixel area value of the boundary area, a diameter inversion operation is performed using a pre-established diameter calibration model to obtain the nugget diameter of the cross steel bar weld. The parameters of the diameter calibration model are calibrated according to the steel bar diameter and welding parameters.

3. The method according to claim 2, characterized in that The step of identifying the boundary area of ​​the cross steel bar welds from the processed thermal radiation image comprises: Identifying, by temperature threshold segmentation, all continuous pixel regions in the processed thermal radiation image whose temperature values ​​are greater than a preset temperature threshold; Establishing a topological relationship graph representing pixel connectivity based on the continuous pixel region, converting a boundary node set in the topological relationship graph into a binary image matrix through coordinate mapping, and generating an initial boundary mask; Performing hole filling processing on the initial boundary mask to obtain a filled boundary mask; performing edge smoothing on the filled boundary mask to obtain a smoothed boundary mask; Based on the smoothed boundary mask, the temperature gradient values ​​of the pixels corresponding to the boundary nodes are calculated, and the pixels with temperature gradient values ​​greater than or equal to the preset temperature gradient threshold are screened to determine the boundary area of ​​the cross-rebar weld.

4. The method according to claim 3, characterized in that The method of calculating the temperature gradient values ​​of pixels corresponding to the boundary nodes based on the smoothed boundary mask and screening pixels whose temperature gradient values ​​are greater than or equal to a preset temperature gradient threshold to determine the boundary area of ​​the cross-rebar weld includes: Based on the smoothed boundary mask, the surface temperature of the boundary nodes in the thermal radiation image is obtained. Combined with the surface temperature of the cross-rebar welds in another thermal radiation image, the temperature gradient value of the corresponding pixel of the boundary node is calculated pixel by pixel along the boundary normal direction, where the boundary normal direction is determined by the coordinates of the adjacent nodes in the topological relationship graph; Obtain the temperature gradient threshold corresponding to the steel bar diameter from the preset configuration table; Pixels with temperature gradient values ​​greater than or equal to the preset temperature gradient threshold are screened to determine the boundary area of ​​the cross-rebar welds.

5. The method according to claim 2, characterized in that According to the pixel area value of the boundary area, a diameter inversion operation is performed using a pre-established diameter calibration model to obtain the nugget diameter of the cross steel bar weld, wherein the parameters of the diameter calibration model are obtained by calibration according to the steel bar diameter and welding parameters, including: According to the steel bar diameter and welding parameters, the parameters of the diameter calibration model are matched from the preset calibration library; According to the pixel area value of the boundary area, combined with the welding position compensation factor, a diameter inversion operation is performed using a pre-established diameter calibration model to obtain the weld core diameter of the cross steel bar weld. The welding position compensation factor is dynamically calculated through the electrode offset.

6. The method according to claim 2, characterized in that The preprocessing of the thermal radiation image to obtain a processed thermal radiation image includes: Separating thermal signals of different frequency bands in the thermal radiation image to remove environmental interference, extract information of the weld nugget edge, and obtain a filtered thermal radiation image; Based on the background temperature value of the non-welding area around the weld point in the filtered thermal radiation image, a sensitivity correction value is calculated to eliminate the radiation error caused by the oxidation of the steel bar surface, and a thermal radiation image with the radiation error eliminated is obtained; Adaptive histogram equalization is performed on the core area of ​​the solder joint in the thermal radiation image after the radiation error is eliminated to obtain a processed thermal radiation image.

7. The method according to claim 6, characterized in that The step of performing adaptive histogram equalization on the solder joint core area in the thermal radiation image after the radiation error is eliminated to obtain a processed thermal radiation image includes: Taking the highest temperature point in the thermal radiation image as the center, a circular area with the diameter of the steel bar as the diameter is defined as the core area of ​​the weld; In the core area of ​​the solder joint, a preset pixel sliding window is used to traverse all pixels and a probability distribution function of the temperature value in the window is calculated; Establishing a mapping relationship from temperature value to equalized grayscale value according to the probability distribution function; Transforming the temperature values ​​of all pixels in the window using the mapping relationship; Bilinear interpolation is used to fuse the transformation results of adjacent windows, and dynamic range compression processing is performed on the transformed temperature value of the core area of ​​the solder joint to suppress the over-enhancement area, thereby obtaining a processed thermal radiation image.

8. A steel mesh multi-weld point quality inspection system based on infrared thermal imaging, characterized in that: include: An acquisition module is used to synchronously acquire, through an infrared thermal imager array, thermal radiation images of the surface temperatures of all cross-rebar welds obtained by the resistance welding process in the steel mesh during the time interval between the end of the resistance welding process and the start of the self-tensioning operation. The cross-rebar welds are welds obtained by the resistance welding process in the steel mesh. an identification module for identifying, for each cross-rebar weld point, a fusion morphology of the cross-rebar weld point based on the thermal radiation image; A generation module, configured to generate a quality inspection result of multiple weld points of the steel mesh based on the fusion morphology of all cross-rebar weld points, wherein the quality inspection result includes a geometric judgment result for indicating whether the fusion morphology meets a preset morphology requirement; The early warning module is used to perform adaptive early warning using a corresponding early warning method according to the quality detection result, so as to adjust the welding parameters used in the next resistance welding process according to the early warning information.

Images