aqora

Reduction of interference fringes in absorption imaging of cold atom cloud using eigenface method

Xiaolin Li, 柯敏, 颜波 +1 more

Chinese Optics LettersJan 1, 20070 citations

Fringe removal algorithms for atomic absorption images: A survey

Gaoyi Lei, Chencheng Tang, Yueyang Zhai

The fringe noises disrupt the precise measurement of the atom distribution in the process of the absorption images. The fringe removal algorithms have been proposed to reconstruct the ideal reference images of the absorption images to remove the fringe noises. However, the focus of these fringe removal algorithms is the association of the fringe removal performance with the physical systems, leaving the gap to analyze the workflows of different fringe removal algorithms. This survey reviews the fringe removal algorithms and classifies them into two categories: the image-decomposition based methods and the deep-learning based methods. Then this survey draws the workflow details of two classical fringe removal algorithms, and conducts experiments on the absDL ultracold image dataset. Experiments show that the singular value decomposition (SVD) method achieves outstanding performance, and the U-net method succeeds in implying the image inpainting idea. The main contribution of this survey is the interpretation of the fringe removal algorithms, which may help readers have a better understanding of the research status. Codes in this survey are available at https://github.com/leigaoyi/Atomic_Fringe_Denoise .The fringe noises disrupt the precise measurement of the atom distribution in the process of the absorption images. The fringe removal algorithms have been proposed to reconstruct the ideal reference images of the absorption images to remove the fringe noises. However, the focus of these fringe removal algorithms is the association of the fringe removal performance with the physical systems, leaving the gap to analyze the workflows of different fringe removal algorithms. This survey reviews the fringe removal algorithms and classifies them into two categories: the image-decomposition based methods and the deep-learning based methods. Then this survey draws the workflow details of two classical fringe removal algorithms, and conducts experiments on the absDL ultracold image dataset. Experiments show that the singular value decomposition (SVD) method achieves outstanding performance, and the U-net method succeeds in implying the image inpainting idea. The main contribution of this survey is the interpretation of the fringe removal algorithms, which may help readers have a better understanding of the research status. Codes in this survey are available at https://github.com/leigaoyi/Atomic_Fringe_Denoise .

Chinese Physics BNov 8, 20211 citation

Noise Suppression and Image Enhancement in Cold-Atom Absorption Imaging

Peiliang Liu, Chaohong Lee, Pengcheng Zheng +7 more

Jan 1, 20240 citations

Deep Learning for Absorption-Image Analysis

Jacob Morrey, I. Peterson, Robert H. Leonard +4 more

The quantum state of ultracold atoms is often determined through measurement of the spatial distribution of the atom cloud. Absorption imaging of the cloud is regularly used to extract this spatial information. Accurate determination of the parameters which describe the spatial distribution of the cloud is crucial to the success of many ultracold atom applications. In this work, we present modified deep learning image classification models for image regression. To overcome challenges in data collection, we train the model on simulated absorption images. We compare the performance of the deep learning models to least-squares techniques and show that the deep learning models achieve accuracy similar to least-squares, while consuming significantly less computation time. We compare the performance of models which take a single atom image against models which use an atom image plus other images that contain background information, and find that both models achieved similar accuracy. The use of single image models will enable single-exposure absorption imaging, which simplifies experiment design and eases imaging hardware requirements.The quantum state of ultracold atoms is often determined through measurement of the spatial distribution of the atom cloud. Absorption imaging of the cloud is regularly used to extract this spatial information. Accurate determination of the parameters which describe the spatial distribution of the cloud is crucial to the success of many ultracold atom applications. In this work, we present modified deep learning image classification models for image regression. To overcome challenges in data collection, we train the model on simulated absorption images. We compare the performance of the deep learning models to least-squares techniques and show that the deep learning models achieve accuracy similar to least-squares, while consuming significantly less computation time. We compare the performance of models which take a single atom image against models which use an atom image plus other images that contain background information, and find that both models achieved similar accuracy. The use of single image models will enable single-exposure absorption imaging, which simplifies experiment design and eases imaging hardware requirements.

arXiv:2506.04517Jun 4, 20250 citations

Advances in Quantum Machine Learning and Deep Learning for Image Classification: A Survey

Ruba Kharsa, Ahmed Bouridane, Abbes Amira

NeurocomputingSep 28, 202356 citations

Absorption imaging of a single atom

Erik W. Streed, Andreas Jechow, B. G. Norton +1 more

arXiv:1201.5280Nature CommunicationsJul 3, 201263 citations

Deep Learning Method for Image Processing in Cold Atom Experiments

Abstract

Related papers

Deep Learning Method for Image Processing in Cold Atom Experiments

Abstract

Related papers