Modes recognition in high-power fiber laser by convolutional neural networks

Jun Li; Hongye Li; Xiaofan Zhao; Zefeng Wang

doi:10.1117/12.2598653

6 June 2021 Modes recognition in high-power fiber laser by convolutional neural networks

Jun Li, Hongye Li, Xiaofan Zhao, Zefeng Wang

Proceedings Volume 11849, Fourth International Symposium on High Power Laser Science and Engineering (HPLSE 2021); 118490J (2021) https://doi.org/10.1117/12.2598653
Event: Fourth International Symposium on High Power Laser Science and Engineering (HPLSE 2021), 2021, Suzhou, China

Abstract

In this paper, mode distribution in large-mode-area (LMA) 25/400 fiber was investigated while attempts to recognize and sort different modes with their combination were carried out on a CNN net via Tensorflow. VGG16 model was chosen as the backbone net through several test to increase precision. The model was trained on a dataset including 6000 pictures in 15 categories. And the final accuracy was up to 0.98. It indicates that recognizing modes in high power fiber laser system based on a CNN net was a feasible plan in the mode control assignment.

Citation Download Citation

Jun Li, Hongye Li, Xiaofan Zhao, and Zefeng Wang "Modes recognition in high-power fiber laser by convolutional neural networks", Proc. SPIE 11849, Fourth International Symposium on High Power Laser Science and Engineering (HPLSE 2021), 118490J (6 June 2021); https://doi.org/10.1117/12.2598653

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available