Vision-based laser cutting wall loss prevention method

Abstract

The present application relates to the technical field of laser cutting, and particularly relates to a laser cutting wall loss prevention method based on vision. The technical scheme comprises the following steps: fixing components and an aluminum framework and setting a to-be-processed pattern, initializing laser processing parameters, determining seed parameters of a half-cut process, executing first laser processing based on the initialized parameters, stopping laser light emission after a set number of processing is completed, taking a photo of a cutting path through a high-magnification vision camera, processing the photographed image and extracting cutting path information using an image processing algorithm, and judging whether cutting is completed according to the image analysis result. The present application combines a vision algorithm with dynamic control, effectively solves the wall damage prevention problem in the laser cutting process without relying on expensive equipment, and takes into account cutting accuracy and processing efficiency, and is suitable for micro devices and other scenes with high requirements for processing precision and damage prevention.

Description

Technical Field

[0001] This invention relates to the field of laser cutting technology, and in particular to a vision-based method for preventing wall damage during laser cutting. Background Technology

[0002] With the rapid development of the 3C and semiconductor industries, the size and volume of devices continue to shrink, making the precision requirements for preventing wall damage during laser cutting increasingly stringent. Currently, existing industry-specific methods for preventing wall damage have significant limitations:

[0003] The passive damage prevention method of adding filler is limited by the trend of miniaturization of device size and can no longer meet the damage prevention requirements in confined spaces;

[0004] Light leakage detection by adding acoustic, optical, and photoelectric sensors (such as spectral devices) to the bottom is not only expensive, but also requires a lot of supporting verification work and has poor adaptability.

[0005] The solution that uses coaxial vision to perceive the completion status of cutting and light leakage lacks a dynamic control mechanism. When the cutting path is only partially completed, it is impossible to determine the subsequent operation strategy, which can easily lead to damage to the wall.

[0006] Therefore, there is an urgent need for a laser cutting wall-cutting loss prevention technology that can dynamically adjust the laser trajectory and power, intelligently predict the detection timing, and take into account processing efficiency, in order to solve the defects of existing solutions in micro-device processing. Summary of the Invention

[0007] The purpose of this invention is to address the problem of preventing damage to the wall during laser cutting in the background art, and to propose a vision-based method for preventing wall damage during laser cutting.

[0008] The technical solution of this invention: a vision-based laser cutting method for preventing wall damage, comprising the following steps:

[0009] S1. Fix the components and aluminum frame and set the drawing to be processed;

[0010] S2. Initialize laser processing parameters and determine seed parameters for the half-cutting process;

[0011] S3. Perform the first laser processing based on the initialization parameters;

[0012] S4. After completing the set number of processing cycles, the laser stops emitting light.

[0013] S5. Take pictures of the cut path using a high-magnification vision camera;

[0014] S6. Use image processing algorithms to process the captured image and extract the cutting information;

[0015] S7. Determine whether the cutting is complete based on the image analysis results;

[0016] S8. If the cutting is not completed, dynamically adjust the cutting trajectory and laser power;

[0017] S9. Implement time-sensitivity control to optimize processing efficiency;

[0018] S10. Save the optimized parameters as initialization parameters for the next round;

[0019] S11. After cutting, the material is fed through an automated system.

[0020] Optionally, step S1, which involves fixing the components and aluminum frame and setting the graphic to be processed, specifically includes: placing the components and aluminum frame in the corresponding carrier and fixing them, and setting the style and size parameters of the graphic to be processed in the canvas.

[0021] Optionally, in step S2, the laser processing parameters are initialized, and the seed parameters for the half-cutting process are determined. Specifically, the process personnel adjust the process parameters for the cutting trajectory. The process parameters include laser power, speed, focal position, and number of processing times to ensure that the aluminum skeleton has no cut-off parts and is undamaged under the process parameters. These process parameters are used as seed parameters for subsequent algorithm iteration and optimization.

[0022] Optionally, step S5, which involves taking pictures of the cutting path with a high-magnification vision camera, specifically includes: using a coaxial or rangefinder high-magnification vision camera to take pictures of the cutting path with the aid of auxiliary light, ensuring that the cutting path is clearly visible and capturing the cutting path area that was cut prematurely due to inconsistent material height.

[0023] Optionally, step S6, which involves processing the captured image using an image processing algorithm and extracting the cutting path information, specifically includes:

[0024] Extract the ROI of the cutting track area to improve processing speed;

[0025] Binarize the image to enhance contrast;

[0026] Perform Blob analysis to remove high-noise points and extract cutting path information;

[0027] The center cutting line is fitted based on the cutting information on both sides. If there are already cut areas, they are removed by fitting.

[0028] Optionally, step S7, which determines whether the cutting is complete based on the image analysis results, specifically includes:

[0029] If the image algorithm detects that the cutting path is completely transparent, it determines that the cutting is complete.

[0030] If there is an opaque area in the cutting path, the cutting is considered incomplete.

[0031] Optionally, the dynamic adjustment of the cutting trajectory and laser power in step S8 specifically includes:

[0032] The cutting trajectory is dynamically adjusted by performing matrix transformation based on the trajectory array and spatial coordinate system information output by the image algorithm to generate canvas trajectory information to update the uncut trajectory.

[0033] The laser power is dynamically adjusted. Based on the image analysis results, the laser power output is reduced by 10% to 90%. For aluminum parts with a thickness of 0.07 to 0.09 mm, the power needs to be reduced after the initial cutting depth of 0.04 to 0.06 mm to avoid overcutting and damaging the carbon steel underneath.

[0034] Optionally, step S9 includes time-based control to optimize processing efficiency, specifically including:

[0035] Set a threshold for the number of processing steps. If the current parameters do not complete the non-destructive cutting within the threshold number of steps, the cutting parameters for the next product will be automatically optimized.

[0036] If the second cutting power reduction of 60% and the third reduction of 80% still fail to complete the cut, adjust the power reduction to 40% for the second cut and 80% for the third cut to achieve non-destructive cutting within 3 cuts.

[0037] Optionally, saving the optimized parameters as initialization parameters for the next round in step S10 specifically includes: saving the optimized parameters from the previous cycle; as the number of cuts increases, removing abnormal data that deviates from the average parameter value of the same batch by more than 20%. For example, for the same product, if the processing is completed in 5 out of 10 times, the abnormal value appears outside this range. If the processing number is 6 times, a deviation of 20% from the standard is still considered to meet the requirements, but if the processing number is 7 times, a deviation of 40% from the benchmark is considered abnormal data and is treated as such. If the number of processing times is too few, the calculation is performed in the same way.

[0038] Compared with the prior art, this application includes at least one of the following beneficial technical effects:

[0039] This invention overcomes the limitations of passive wall-mounting damage prevention methods such as adding fillers due to the confined space of devices, making it suitable for micro-device fabrication scenarios. It eliminates the need for expensive detection equipment such as spectral analysis, reducing equipment costs and the complexity of supporting verification work. It can dynamically adjust the laser trajectory and optimize the processing path of uncut areas based on visual algorithm analysis results, improving cutting accuracy.

[0040] This invention dynamically adjusts the laser power according to the actual conditions of the cutting path, avoiding overcutting and damage to the underlying material due to excessive power, thus achieving non-destructive cutting. Through aging control and iterative parameter optimization, it can also dynamically adjust processing parameters to improve cutting efficiency and reduce the number of processing steps. This invention features an intelligent cutting status judgment mechanism, which, combined with visual inspection, can accurately determine whether the cutting is complete, avoiding invalid or insufficient processing.

[0041] This invention combines visual algorithms with dynamic control to effectively solve the problem of wall damage prevention during laser cutting without relying on expensive equipment. It also balances cutting accuracy and processing efficiency, making it suitable for scenarios with high requirements for processing precision and damage prevention, such as micro-devices. Attached Figure Description

[0042] Figure 1 This is a flowchart of a vision-based laser cutting method for preventing wall damage. Detailed Implementation

[0043] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0044] Example

[0045] like Figure 1 As shown, a vision-based laser cutting method for preventing wall damage according to the present invention includes the following steps:

[0046] Step S1: Place the components and the 0.08mm thick aluminum frame in the corresponding carrier and fix them in place. Set the graphic to be processed in the canvas. Fixing with the carrier ensures the stability of the components and aluminum frame during processing, avoiding deviations in the cutting trajectory due to displacement. The preset processing graphic in the canvas provides a precise path reference for subsequent laser processing, ensuring the accuracy of the initial processing.

[0047] Step S2: Initialize laser processing parameters. For the corresponding cutting trajectory, the process personnel adjust the process parameters, such as laser power, speed, laser focus position, and number of processing operations. This ensures that no areas of the skeleton are cut under the current cutting parameters, guaranteeing no damage. Subsequent algorithms can use these seed parameters for iterative optimization. Pre-tuned seed parameters ensure that the initial cut will not cut through the skeleton and cause no damage, providing a safe and reliable foundation for subsequent algorithm iteration and optimization, and avoiding the risk of wall damage due to unreasonable initial parameters.

[0048] Step S3: The first laser processing begins, and the aluminum frame is cut based on the initialized parameters. Processing is started with verified seed parameters, and the initial cut is completed while ensuring safety against the wall, providing a practical cutting sample for subsequent visual inspection and parameter adjustment.

[0049] Step S4: After the processing cycle is completed, the laser stops emitting light. Stopping light emission after a preset number of cycles avoids unnecessary continuous processing that could lead to power accumulation and damage, while also providing a stable, laser-free environment for visual inspection.

[0050] Step S5: Take pictures of the cut path using a coaxial or rangefinder high-magnification vision camera to ensure the cut path is clearly visible. Due to the inconsistency in material height, some cut paths may be completed prematurely. The camera image can clearly capture the completed areas with auxiliary light. The high-magnification camera combined with auxiliary light can clearly capture the details of the cut path, especially distinguishing the prematurely completed areas caused by the inconsistency in material height, providing high-quality data for subsequent image analysis and ensuring detection accuracy.

[0051] Step S6: Input the captured image. The image processing algorithm can be Halcon or OpenCV.

[0052] The ROI of the cutting channel area is extracted to improve the processing speed of the algorithm.

[0053] Binarization of the corresponding images enhances contrast.

[0054] Blob analysis removes high noise levels and extracts cutting path information.

[0055] Cutting track information processing: a) Fit the center cutting line (i.e., the laser cutting trajectory) based on the cutting track information on both sides; b) If some parts have already been cut, remove the corresponding areas through fitting. This provides a precise basis for trajectory adjustment and avoids reprocessing the completed parts.

[0056] Step S7: Determine if the cutting is complete:

[0057] a. Through the above image processing algorithm, the cutting path is detected to be clear. The cutting work has been completed, and the judgment result is yes.

[0058] b. If there are still opaque areas on the cutting path, the result is negative. Objective image-based judgment replaces human experience, accurately identifying uncut areas and avoiding undercutting or overcutting due to misjudgment, thus improving the stability of cutting quality.

[0059] Step S8: a) Dynamic adjustment of the cutting trajectory: Based on the trajectory array and spatial coordinate system information output by the image algorithm, a matrix transformation is performed to obtain the canvas trajectory information, completing the dynamic transformation of the uncut trajectory. Dynamic trajectory adjustment ensures that the laser only acts on the uncut area, reducing invalid paths and improving processing efficiency. b) Dynamic adjustment of laser power: Based on the image algorithm processing results, the laser power output is reduced by 60%. Reducing the power can avoid overcutting (e.g., after a 0.08mm aluminum part has been cut to 0.05mm, low power can prevent damage to the carbon steel below), fundamentally solving the problem of wall damage prevention.

[0060] Step S9: Time-based control. This stage primarily aims to improve cutting efficiency. For example, if the cutting is still not completed after the second cutting power is reduced by 60% and the third cutting power is reduced by 80%, the cutting parameters for the next product will be self-optimized through this step. The second power might be reduced by 40%, and the third by 80%, achieving non-destructive cutting in three attempts. Alternatively, we can specify that cutting can be completed in two attempts, which requires the algorithm to self-update and optimize. By dynamically adjusting parameters based on historical processing data, the processing cycle is shortened while ensuring non-destructive cutting, significantly improving overall processing efficiency and adapting to the needs of mass production.

[0061] Step S10: Save the optimized parameters from the previous cycle and use them as the initialization parameters for the next round of cutting. As the number of cuts increases, eliminate abnormal data and use the average of multiple parameter values ​​as the initialization parameters for the next round of cutting. Accumulating optimal process data through parameter iteration and using the average value after eliminating outliers reduces random errors, making the initialization parameters for subsequent processing more accurate, reducing debugging costs, and improving process stability. As the number of cuts increases, eliminate abnormal data that deviates from the average parameter value of the same batch by more than 20%. For example, for the same product, if processing is completed in 5 out of 10 cycles, an abnormal value outside this range is considered acceptable. For example, if processing is completed in 6 cycles, a 20% deviation from the standard is still considered acceptable, but if processing is completed in 7 cycles, a 40% deviation from the baseline is considered abnormal data. If the number of processing cycles is too low, calculate using the same method.

[0062] Step S11: Complete the cutting of this aluminum frame and feed it through an automated system. Automated feeding reduces manual intervention, lowers the risk of positional deviation caused by human operation, and enables continuous production, further improving overall processing efficiency.

[0063] This invention dynamically adjusts the laser trajectory, using visual algorithms to analyze the size and shape of incomplete cuts to generate a new trajectory. This ensures the laser only acts on uncut areas, avoiding accidental damage to completed cut areas or the opposite wall. Simultaneously, it dynamically adjusts the laser power, reducing it by 10% to 90% depending on the cutting situation to prevent light leakage and damage to the opposite wall due to excessive power. This is particularly suitable for scenarios where fragile materials like carbon steel lie beneath 0.08mm thick aluminum parts. A coaxial or off-axis high-magnification vision camera is used to photograph the cut path. Combined with image processing algorithms such as Halcon or OpenCV, the cut path information is accurately extracted through steps such as ROI extraction, binarization, and blob analysis. The center cutting line is fitted, and completed areas are removed, providing an accurate basis for trajectory adjustment and reducing cutting deviations.

[0064] The algorithm of this invention includes time-based control, dynamically adjusting the power based on the cutting situation of each processing operation, and reducing the number of processing operations through parameter self-optimization (e.g., from 3 to 2). Simultaneously, dynamic trajectory adjustment avoids reprocessing of completed parts, significantly improving overall processing efficiency. It eliminates the need for expensive detection equipment such as spectral analysis, reducing equipment investment costs; it also eliminates the need for extensive verification work, and through iterative parameter optimization, reduces the frequency of manual debugging and operational complexity.

[0065] It's worth noting that by initializing seed parameters and iteratively optimizing them, the optimized parameters are saved as the initial parameters for the next round of cutting. After removing abnormal data, the average of multiple parameter values ​​is used as the initial value, making the processing parameters more stable and reliable, and ensuring the consistency of cutting quality. The system can intelligently predict when to stop light and take pictures, stopping light emission and taking pictures after the processing cycle is completed. Combined with image algorithms, it determines whether the cutting is complete, avoiding invalid or insufficient processing and improving the intelligence level of the cutting process.

[0066] The above specific embodiments are merely several optional embodiments of the present invention. Based on the technical solutions of the present invention and the relevant teachings of the above embodiments, those skilled in the art can make various alternative improvements and combinations to the above specific embodiments.

Claims

1. A vision-based laser cutting method for preventing wall damage, characterized in that, Includes the following steps: S1. Fix the components and aluminum frame and set the drawing to be processed; S2. Initialize laser processing parameters and determine seed parameters for the half-cutting process; S3. Perform the first laser processing based on the initialization parameters; S4. After completing the set number of processing cycles, the laser stops emitting light. S5. Take pictures of the cut path using a high-magnification vision camera; S6. The captured image is processed using an image processing algorithm to extract the cutting path information, specifically including: Extract the ROI of the cutting track area to improve processing speed; Binarize the image to enhance contrast; Perform Blob analysis to remove high-noise points and extract cutting path information; The center cutting line is fitted based on the cutting information on both sides. If there are already cut areas, they are removed by fitting. S7. Determine whether the cutting is complete based on the image analysis results, specifically including: If the image algorithm detects that the cutting path is completely transparent, it determines that the cutting is complete. If there is an opaque area in the cutting path, the cutting is considered incomplete. S8. If cutting is not completed, dynamically adjust the cutting trajectory and laser power, specifically including: The cutting trajectory is dynamically adjusted by performing matrix transformation based on the trajectory array and spatial coordinate system information output by the image algorithm to generate canvas trajectory information to update the uncut trajectory. The laser power is dynamically adjusted. Based on the image analysis results, the laser power output is reduced by 10% to 90%. For aluminum parts with a thickness of 0.07 to 0.09 mm, the power needs to be reduced after the initial cutting depth of 0.04 to 0.06 mm to avoid overcutting and damaging the carbon steel underneath. S9. Implement time-sensitivity control to optimize processing efficiency, specifically including: Set a threshold for the number of processing steps. If the current parameters do not complete the non-destructive cutting within the threshold number of steps, the cutting parameters for the next product will be automatically optimized. If the second cutting power is reduced by 60% and the third power is reduced by 80% and still not completed, adjust to reduce the power by 40% for the second time and 80% for the third time to achieve non-destructive cutting within 3 times; S10. Save the optimized parameters as the initialization parameters for the next round. Specifically, this includes saving the optimized parameters from the previous cycle and removing abnormal data that deviate from the average value of the same batch of parameters by more than 20% as the number of cuts increases. S11. After cutting, the material is fed through an automated system.

2. The vision-based laser cutting method for preventing wall damage according to claim 1, characterized in that, Step S1, which involves fixing the components and aluminum frame and setting the graphic to be processed, specifically includes: placing the components and aluminum frame in the corresponding carrier and fixing them, and setting the style and size parameters of the graphic to be processed in the canvas.

3. The vision-based laser cutting method for preventing wall damage according to claim 1, characterized in that, In step S2, the laser processing parameters are initialized, and the seed parameters for the half-cutting process are determined. Specifically, the process personnel adjust the process parameters for the cutting trajectory. The process parameters include laser power, speed, focal position, and number of processing times to ensure that the aluminum skeleton has no cut-off parts and is undamaged under the process parameters. These process parameters serve as seed parameters for subsequent algorithm iteration and optimization.

4. The vision-based laser cutting method for preventing wall damage according to claim 1, characterized in that, Step S5, which involves taking pictures of the cutting path with a high-magnification vision camera, specifically includes: using a coaxial or rangefinder high-magnification vision camera, and taking pictures of the cutting path with the aid of auxiliary light, to ensure that the cutting path is clearly visible and to capture the cutting path area that was cut prematurely due to inconsistent material height.

Images