Vision Camera Calibration for Precise Laser Welding Targeting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In laser welding processes for battery manufacturing, accurately designating the welding target region is challenging due to variations in vision camera settings, leading to defects and reduced yield in battery modules.
Innovation Solution
A laser welding apparatus with a vision camera that optimizes setting parameters by comparing sample RGB values from a welding target region with reference RGB values, adjusting parameters such as exposure time and gain to ensure accurate identification of the welding target, thereby improving the precision and stability of the welding process.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If vision camera setting parameters are not optimized, then the device complexity is reduced, but the manufacturing precision of the welding target region identification deteriorates
Solution Approach 1:
The system performs preliminary action by capturing a sample image of the welding target region before the actual welding process and comparing it with a reference image to pre-optimize vision camera parameters. This preliminary parameter optimization ensures that the vision system is properly calibrated for accurate welding target identification, resolving the contradiction by establishing proper settings in advance rather than during the welding process itself.
Solution Approach 2:
The system implements feedback by comparing the sample image obtained from the vision camera with a reference image, analyzing the difference in RGB values, and automatically adjusting the vision camera parameters based on this comparison. This closed-loop feedback mechanism enables automatic optimization of imaging parameters to achieve precise welding target region identification without requiring complex manual calibration procedures.
2Manufacturing precision
If vision camera parameters are adjusted to improve welding target identification accuracy, then the manufacturing precision improves, but the productivity decreases due to additional parameter optimization steps
Solution Approach 1:
The parameter optimization is performed as a preliminary step before mass production welding begins. By establishing the optimal vision camera parameters in advance through sample image comparison, the system ensures high welding target identification precision while avoiding the need to repeatedly adjust parameters during production, thus minimizing the impact on overall productivity.
Solution Approach 2:
The system implements self-service by automatically comparing the sample image with the reference image and autonomously determining the optimal vision camera parameters without requiring operator intervention. This automated parameter optimization reduces the time and labor required for setup, thereby minimizing the impact on productivity while achieving high manufacturing precision.
3Measurement precision
If the vision camera captures detailed images of the welding target region, then the measurement precision improves, but the loss of time increases due to image processing and parameter adjustment
Solution Approach 1:
The system performs the time-consuming image capture and parameter optimization as a preliminary action before the actual welding production begins. By capturing the sample image and completing parameter adjustment in advance, the system achieves high measurement precision for welding target identification without causing time loss during the production process itself.
Solution Approach 2:
The system uses a reference image as a template or copy of the ideal welding target region appearance. By comparing the sample image against this reference copy, the system can quickly evaluate whether parameter adjustment is needed and make precise adjustments, reducing the time required for measurement and optimization while maintaining high precision.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A laser welding apparatus includes a stage on which a first structure and a second structure, targets of a laser welding process, are seated, a vision camera located above the stage and obtaining a sample image indicating a welding target region in which the first structure and the second structure come into contact, and a controller adjusting setting parameters of the vision camera. The controller obtains a sample RGB value from each of a plurality of regions of interest selected from the sample image, compares the sample RGB value with a reference RGB value obtained from each of reference regions of interest defined in the same location as the plurality of regions of interest in a predetermined reference image, to adjust at least one of the setting parameters, and sets the vision camera.