Neural-network-assisted in situ processing monitoring by speckle pattern observation

TL;DR

This paper introduces a neural-network-based method for real-time monitoring of laser processing by analyzing speckle patterns, enabling accurate depth prediction and material identification without complex analytical models.

Contribution

It presents a novel approach combining high-speed speckle pattern observation with deep learning to monitor laser processing in situ, improving accuracy and versatility.

Findings

Predicted ablation depth with 2 micron uncertainty.

Successfully identified material during processing.

Enabled real-time monitoring without analytical models.

Abstract

We propose a method to monitor the progress of laser processing using laser speckle patterns. Laser grooving and percussion drilling were performed using femtosecond laser pulses. The speckle patterns from a processing point were monitored with a high-speed camera and analyzed with a deep neural network. The deep neural network enabled us to extract multiple information from the speckle pattern without a need for analytical formulation. The trained neural network was able to predict the ablation depth with an uncertainty of 2 \micron, as well as the material under processing, which will be useful for composite material processing.