Transfer Learning of Reconstructing Imaging From Speckles Generated by Different Spatial Coherence Light
Open Access
- 21 October 2024
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 12, 153911-153921
Abstract
In recent years, deep learning has been effectively used to reconstruct speckled images created by the scattering media. However, when a network with previous data training is used to test new data, it usually exhibits poor performance due to significant differences in new data that differ in scattered media and data types. Retraining the network and collecting a big amount of new data are usually required to enhance performance, but this comes with a high cost and time expense. This article proposes a transfer learning approach to address this issue. We constructed an experimental setup for capturing infrared speckled images, which includes a laser module with a 4f resonant cavity structure capable of adjusting the aperture size of spatial filters. Speckled images are collected and preprocessed at different aperture sizes, then input into a convolutional neural network for training. Transfer learning is employed to study the reconstruction of speckled images between different aperture sizes. Subsequently, to recreate speckle patterns, a speckle reconstruction network named TCGAN (Transformer_CNN_GAN) is proposed. This network is based on pre-training TCGAN with an enormous quantity of data and fine-tuning its model with new data so that it can be adapted quickly with a small sample size. The experimental findings show that TCGAN may achieve good reconstruction without too much new data by means of transfer learning. Moreover, pretraining the network with more complex datasets helps enhance the network’s transfer learning capability. Therefore, conducting research on optical field imaging and restoration methods based on transfer learning holds significant theoretical and practical significance.Keywords
Funding Information
- Opening Funding of the National Key Laboratory of Electromagnetic Space Security
- Shenyang Ligong University’s Scientific Research Support Plan for Introducing High-Level Talents
This publication has 30 references indexed in Scilit:
- Reconstructing Images Through Multimode Fibers From the Up-Conversion Speckle Patterns via Deep LearningIEEE Access, 2023
- CRMSNet: A deep learning model that uses convolution and residual multi‐head self‐attention block to predict RBPs for RNA sequenceProteins: Structure, Function, and Bioinformatics, 2023
- Single Frame Atmospheric Turbulence Mitigation: A Benchmark Study and a New Physics-Inspired Transformer Model2022
- Accelerate adversarial training with loss guided propagation for robust image classificationInformation Processing & Management, 2022
- Experimental study on frequency doubling of Q-switched partially coherent laserOptical Review, 2022
- A survey of deep learning approaches to image restorationNeurocomputing, 2022
- Mutual Transfer Learning of Reconstructing Images Through a Multimode Fiber or a Scattering MediumIEEE Access, 2021
- End-to-End Object Detection with Transformers2020
- Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network2017
- The Correlation Coefficient: An OverviewCritical Reviews in Analytical Chemistry, 2006