基于傅里叶成像网络的全息粒子成像技术研究

doi:10.7517/issn.1674-0475.231026

影像科学与光化学 ›› 2024, Vol. 42 ›› Issue (1): 1-8.DOI: 10.7517/issn.1674-0475.231026

基于傅里叶成像网络的全息粒子成像技术研究

李宇杰¹, 程伟哲², 余乐², 王华英¹, 王丹¹, 李学燕¹, 李伯仕¹

1. 河北工程大学数理科学与工程学院, 河北邯郸 056038;
2. 北京理工大学深圳研究院, 广东深圳 518057

收稿日期:2023-10-27 修回日期:2023-12-11 发布日期:2024-02-21
通讯作者: 余乐
基金资助:
深圳市科技计划项目资助(JCYJ20220530164808019, JCYJ20230807113559010)

A Study on Holographic Particle Imaging Based on Fourier Imaging Networks

LI Yujie¹, CHENG Weizhe², YU Le², WANG Huaying¹, WANG Dan¹, LI Xueyan¹, LI Boshi¹

1. School of Mathematics and Physics Science and Engineering, Hebei University of Engineering, Handan 056038, Hebei, P.R. China;
2. Shenzhen Research Institute, Beijing Institute of Technology, Shenzhen 518057, Guangdong, P.R. China

Received:2023-10-27 Revised:2023-12-11 Published:2024-02-21

摘要/Abstract

摘要： 数字全息图重建可以获得物体的三维信息,在医疗、环保、化工等需要测量颗粒大小和三维信息的相关研究有着重要的科学意义和使用价值。粒子全息定位通常有计算复杂和重建精度不足等问题,为了解决相关问题,我们采用一种基于傅里叶成像网络的方法,利用傅里叶变换的特性,能够有效提取全息图的频率特征,结合深度学习技术,通过可学习的滤波器和全局感受野处理输入数据的空间频率,实现对全息图快速和准确的处理,不仅提高了数据处理效率,还具备了更强大的特征提取能力,从而实现对粒子的精确定位。本文通过实验和仿真两方面验证,并与Dense_U_net网络对比。结果显示,基于傅里叶成像网络的方法在定位精度和速度上都有显著的提升。并通过实验验证了该方法可行性。

关键词: 数字全息, 深度学习, 傅里叶变换, 粒子定位

Abstract: Digital holographic reconstruction can obtain three-dimensional information of objects, and it has significant scientific significance and practical value in medical, environmental protection, chemical engineering, and other fields that require measurement of particle size and three-dimensional information. The particle holographic localization often faces issues of computational complexity and insufficient reconstruction accuracy. To address these issues, we adopted a method based on the Fourier imaging network. By leveraging the characteristics of the Fourier transform, we were able to effectively extract frequency features from the hologram. Combined with deep learning technology, the network utilized learnable filters and a global receptive field to process the spatial frequencies of input data. This approach not only enhanced data processing efficiency but also endowed the system with superior feature extraction capabilities, thereby achieving precise particle localization. In this paper, we validated our approach through both experimental and simulation studies and compared it with the Dense_U_net network. The results show that the method based on the Fourier Imaging Network significantly improved both localization accuracy and speed.

Key words: digital holography, deep learning, Fourier transform, particle localization

李宇杰, 程伟哲, 余乐, 王华英, 王丹, 李学燕, 李伯仕. 基于傅里叶成像网络的全息粒子成像技术研究[J]. 影像科学与光化学, 2024, 42(1): 1-8.

LI Yujie, CHENG Weizhe, YU Le, WANG Huaying, WANG Dan, LI Xueyan, LI Boshi. A Study on Holographic Particle Imaging Based on Fourier Imaging Networks[J]. IMAGING SCIENCE AND PHOTOCHEMISTRY, 2024, 42(1): 1-8.

参考文献

[1] Yi F, Moon I, Javidi B. Automated red blood cells extraction from holographic images using fully convolutional neural networks[J]. Biomedical Optics Express, 2017, 8(10):4466-4479.
[2] Wu X C, Li X W, Yao L C, et al. Accurate detection of small particles in digital holography using fully convolutional networks[J]. Applied Optics, 2019, 58(34):G332-G344.
[3] Hussain R, Alican Noyan M, Woyessa G, et al. An ultra-compact particle size analyser using a CMOS image sensor and machine learning[J]. Light:Science&Applications, 2020, 9(1):21.
[4] Zhang G, Guan T, Shen Z Y, et al. Fast phase retrieval in off-axis digital holographic microscopy through deep learning[J]. Optics Express, 2018, 26(15):19388-19405.
[5] Sherstinsky A. Fundamentals of recurrent neural network (RNN) and long short-term memory (LSTM) network[J]. Physica D:Nonlinear Phenomena, 2020, 404:132306.
[6] Rivenson Y, Wu Y C, Ozcan A. Deep learning in holography and coherent imaging[J]. Light:Science&Applications, 2019, 8(1):85.
[7] Rivenson Y, Zhang Y B, Günaydın H, et al. Phase recovery and holographic image reconstruction using deep learning in neural networks[J]. Light:Science&Applications, 2018, 7(2):17141.
[8] Wang K Q, Dou J Z, Qian K M, et al. Y-Net:A one-to-two deep learning framework for digital holographic reconstruction[J]. Optics Letters, 2019, 44(19):4765-4768.
[9] O'Connor T, Anand A, Andemariam B, et al. Deep learning-based cell identification and disease diagnosis using spatio-temporal cellular dynamics in compact digital holographic microscopy[J]. Biomedical Optics Express, 2020, 11(8):4491-4508.
[10] Shimobaba T, Takahashi T, Yamamoto Y, et al. Digital holographic particle volume reconstruction using a deep neural network[J]. Applied Optics, 2019, 58(8):1900-1906.
[11] Wu Y F, Wu J C, Jin S Z, et al. Dense-U-net:Dense encoder-decoder network for holographic imaging of 3D particle fields[J]. Optics Communications, 2021, 493:126970.
[12] Chen H L, Huang L Z, Liu T R, et al. Fourier Imager Network (FIN):A deep neural network for hologram reconstruction with superior external generalization[J]. Light:Science&Applications, 2022, 11(1):254.
[13] Kovachki N, Lanthaler S, Mishra S. On universal approximation and error bounds for Fourier Neural Operators[J]. The Journal of Machine Learning Research, 2021, 22(1):13237-13312.
[14] 曹良才,何泽浩,赵燕,等.面向层结构的角谱传播计算全息算法(特邀)[J].红外与激光工程, 2018, 47(6):11-19.
[15] Li M, Hsu W, Xie X D, et al. SACNN:Self-attention convolutional neural network for low-dose CT denoising with self-supervised perceptual loss network[J]. IEEE Transactions on Medical Imaging, 2020, 39(7):2289-2301.
[16] Amini N, Mostaar A. Deep learning approach for fusion of magnetic resonance imaging-positron emission tomography image based on extract image features using pretrained network (VGG19)[J]. Journal of Medical Signals and Sensors, 2022, 12(1):25.

基于傅里叶成像网络的全息粒子成像技术研究

A Study on Holographic Particle Imaging Based on Fourier Imaging Networks

RichHTML

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 12

编辑推荐

Metrics

本文评价

[1]	朱永月, 周舟, 李艳若, 郭伟, 王默涵, 王道清. 深度学习算法在肝细胞肝癌影像学诊疗领域的研究进展[J]. 影像科学与光化学, 2023, 41(6): 345-349.
[2]	吕刚, 应明亮. 基于3D卷积网络的结肠癌CT图像分类研究[J]. 影像科学与光化学, 2022, 40(5): 1024-1028.
[3]	陈健兵, 葛磊, 文永富. 基于动态散斑干涉的数字全息成像技术研究[J]. 影像科学与光化学, 2022, 40(4): 697-703.
[4]	葛磊, 文永富, 程灏波. 双波长离轴数字全息成像技术研究[J]. 影像科学与光化学, 2021, 39(6): 790-797.
[5]	魏淑玉, 陈欣, 文永富, 程灏波. 数字全息重构图像边缘误差抑制方法研究[J]. 影像科学与光化学, 2021, 39(2): 166-173.
[6]	王倩, 陈欣, 文永富, 杨昊, 王华英, 程灏波. 大数值孔径光学镜约束多级衍射机理[J]. 影像科学与光化学, 2020, 38(2): 181-189.
[7]	吴育民, 段海燕, 文永富, 程灏波, 王谭. 再现图像细节抑制散斑噪声技术研究[J]. 影像科学与光化学, 2018, 36(2): 187-191.
[8]	盛海见, 吴育民, 文永富, 李根, 程灏波. 无透镜数字全息显微成像技术与应用[J]. 影像科学与光化学, 2017, 35(2): 99-105.
[9]	李凤娇, 张启灿, 刘守起, 吴应山. 基于虚拟相位平面的相位展开方法[J]. 影像科学与光化学, 2017, 35(2): 185-192.
[10]	门高夫, Pascal Picart, 王华英, 江夏男, 楼宇丽. 菲涅耳非相干相关数字全息研究进展[J]. 影像科学与光化学, 2017, 35(2): 199-207.
[11]	吴育民, 宋国庆, 冯云鹏, 王钟. 数字全息显微在医学影像中的发展与最新应用[J]. 影像科学与光化学, 2016, 34(1): 23-29.
[12]	赵宝群, 栗军香, 江夏男, 朱巧芬. 像前与像后预放大与数字全息显微成像比较[J]. 影像科学与光化学, 2016, 34(1): 68-74.