Work

2D phase unwrapping by attention-guided deep neural network for optical interferometry

Year: 2025

Type: article

Abstract: Two-dimensional phase unwrapping is an important process in interferometry applications since the unwrapped phase plays a guiding role in the morphological reconstruction of the object. However, the unwrapping phase for each position is affected by the wrap states for other positions. Therefore, it is essential to obtain global information for better phase unwrapping. To this purpose, we propose a novel attention-guided deep neural network that introduces the self-attention mechanism, a popular deep learning technique, to tackle the problem in a high efficient way. We design a dual attention phase unwrapping network (DAPUN), which uses two kinds of complementary self-attention modules to obtain global wrapped states, allowing each positional feature to be directly compared with ones at any other positions. To evaluate the effectiveness of the proposed DAPUN, we first simulate extensive data to pre-train the networks and compare it with several existing phase unwrapping methods. The results show that DAPUN significantly outperforms the previous state-of-the-art. After that, we apply the pre-trained DAPUN to the real cases collected from optical interferometry, demonstrating its generalization capability.

Source: Optics & Laser Technology

DOI: 10.1016/j.optlastec.2025.113358

Authors: Youxing Li, Lingzhi Meng, Donghui Wang, Jiahao Zhang, Libo Yuan

Institutions: Harbin Engineering University, Guilin University of Electronic Technology

Language: English

Topic: Optical measurement and interference techniques

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

Estimated APC paid: $3,230

Funder: Natural Science Foundation of Guangxi Zhuang Autonomous Region