Machine vision-based high-precision and robust focus detection for femtosecond laser machining

Area of Science:

• Optics and Photonics
• Precision Engineering
• Machine Vision

Background:

• Accurate focus detection is critical for high-precision femtosecond laser machining.
• Existing methods may struggle with tilted or curved surfaces, limiting applicability.
• The need for real-time, non-contact focus monitoring in advanced manufacturing.

Purpose of the Study:

• To develop and validate a machine vision-based focus detection method (MVFD) for femtosecond laser machining.
• To achieve simultaneous determination of defocus direction and distance.
• To assess the method's precision and robustness on various sample types.

Main Methods:

• Utilized machine vision to analyze the laser focus pattern.
• Developed algorithms to extract defocus direction and distance from the pattern.
• Tested the method on flat, tilted, and curved surface samples.

Main Results:

• Achieved high precision on flat surfaces with an average error of 0.047 µm and RMSE of 0.055 µm.
• Demonstrated robustness on tilted samples, with minimal impact on accuracy.
• Obtained good performance on curved surfaces, with an average error of 0.093 µm and RMSE of 0.145 µm.

Conclusions:

• The proposed MVFD offers a precise and robust solution for focus detection in femtosecond laser machining.
• The method is easily integrated with existing laser processing equipment.
• Enables high-precision manufacturing over large ranges and complex geometries.