2022 · SPIE Photonics Europe 2022

Machine learning-based high-precision and real-time focus detection for laser material processing systems

Can Polat , Gizem Nuran Yapici , Sepehr Elahi , Parviz Elahi

Research highlights

Fast focus detection for laser processing

Predict focal position from camera images quickly enough for a laser-processing feedback loop.

Why we investigated it

An incorrectly focused beam can fail to process a material as intended, making both precision and inference speed important.

What the work contributes

The study compares CNN and non-CNN approaches using an ordinary lens and CMOS camera.

What the evidence shows

The proposed CNN reports 95% focus-classification accuracy on the tested distance grid and CPU inference above 1,000 Hz.

Build on this work

Use the optical setup, focus grid, and model comparison as a reference for designing a real-time autofocus experiment.

  • Laser processing
  • Autofocus
  • Convolutional neural networks
  • Real-time inference
Abstract
This work explores a real-time and high precision focus finding for the ultrafast laser material processing for a different types of materials. Focus detection is essential for laser machining because an unfocused beam cannot affect the material and, at worst, a destructive effect. Here, we compare CNN and non-CNN-based approaches to focus detection, ultimately proposing a robust CNN model that can achieve high performance when only trained on a portion of the dataset. We use an ordinary lens (11 mm focal length, 0.25 NA) and a CMOS camera. Our robust CNN model achieved a focus prediction accuracy of 95% when identifying focus distances in {-150, -140,...,0,...,150} µm, each step is about 7% of the Rayleigh length, and a high processing speed of 1000+ Hz on a CPU
Cite
@article{polat2022machine,
  title = {Machine learning-based high-precision and real-time focus detection for laser material processing systems},
  author = {Can Polat and Gizem Nuran Yapici and Sepehr Elahi and Parviz Elahi},
  year = 2022,
  journal = {SPIE Photonics Europe 2022},
  doi = {10.1117/12.2624383},
}