WANG Xiang; ZHOU Yishen; ZHANG Xuange; CHEN Xihao

doi:10.7498/aps.74.20250010

Abstract

This paper presents a novel convolutional neural network-based single-pixel imaging method that integrates a physics-driven fusion attention mechanism. By incorporating a module that combines both channel attention mechanism and spatial attention mechanism into a randomly initialized convolutional network, the method utilizes the physical model constraints of single-pixel imaging to achieve high-quality image reconstruction. Specifically, the spatial and channel attention mechanism are combined into a single module and introduced into various layers of a multi-scale U-net convolutional network. In the spatial attention mechanism, we extract the attention weight features of each spatial region of the pooled feature map by using convolution. In the channel attention mechanism, we pool the three-dimensional feature map into a single-channel signal and input it into a two-layer fully connected network to obtain the attention weight information for each channel. This approach not only uses the critical weighting information provided by the attention mechanism in the three-dimensional data cube but also fully integrates the powerful feature extraction capabilities of the U-net network across different spatial frequencies. This innovative method can effectively capture image details, suppress background noise, and improve image reconstruction quality. During the experimental phase, we employ the optical path of single-pixel imaging to acquire bucket signals for two target images, "snowflake" and "basket". By inputting any noisy image into a randomly initialized neural network with attention mechanism, and using the mean square error between simulated bucket signal and actual bucket signal, we physically constrain the convergence of the network. Ultimately, we achieve a reconstructed image that adheres to the physical model. The experimental results demonstrate that under low sampling rate conditions, the scheme of integrating the attention mechanism can not only intuitively reconstruct image details better, but also demonstrate significant advantages in quantitative evaluation metrics such as peak signal-to-noise ratio (PSNR) and structural similarity (SSIM), confirming its effectiveness and potential application in single-pixel imaging.

Keywords:

Authors and contacts

Corresponding author: CHEN Xihao, xi-haochen@163.com

Funds: Project supported by the National Key Research and Development Program of China (Grant No. 2018YFB0504302).

FullText HTML

References

[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]

Cited By

DownLoad: Full-Size Img PowerPoint

DownLoad: Full-Size Img PowerPoint

DownLoad: Full-Size Img PowerPoint

DownLoad: Full-Size Img PowerPoint

DownLoad: Full-Size Img PowerPoint

DownLoad: Full-Size Img PowerPoint

DownLoad: Full-Size Img PowerPoint

map

[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]
[34]
[35]
[36]
[37]
[38]
[39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
[51]
[52]
[53]
[54]
[55]
[56]
[57]
[58]
[59]

[1]	SUN Kangsheng, HAN Chao, QIN Haifeng, GU Tao, LI Wei, YU Cheng. Fast generation algorithm of high-quality holograms based on attention convolutional neural network. Acta Physica Sinica, 2025, 74(8): 084203. doi: 10.7498/aps.74.20241713
[2]	JIANG Tingshuai, RUAN Yirun, LI Hai, BAI Liang, YUAN Yifei, YU Tianyuan. Entropy-based weighted multi-channel convolutional neural network method for node importance assessment. Acta Physica Sinica, 2025, 74(12): . doi: 10.7498/aps.74.20250329
[3]	Wu Yin-Hua, Chong Zhe, Zhu Peng-Fei, Chen Sha-Sha, Zhou Shun. Removal of background white light in coherent-dispersion spectrometer based on convolutional neural network. Acta Physica Sinica, 2025, 74(10): . doi: 10.7498/aps.74.20250090
[4]	Chen Ming-Lai, Ma Cai-Wen, Liu Hui, Luo Xiu-Juan, Feng Xu-Bin, Yue Ze-Lin, Zhao Jing. Fast sampling based image reconstruction algorithm for sheared-beam imaging. Acta Physica Sinica, 2024, 73(2): 024202. doi: 10.7498/aps.73.20231254
[5]	He Zhi-Ye, Zhang Yan-Dong, Tang Chun-Hua, Li Jun-Li, Li Si-Wei, Yu Bin. Analysis of influence of imaging resolution of relay lens on image reconstruction quality in pixel-wise coded exposure imaging technology. Acta Physica Sinica, 2023, 72(2): 024201. doi: 10.7498/aps.72.20221588
[6]	Zhang Jian, Chen Jia-Lin, Chen Xiao-Ran, Mao Tian-Yi, Shen Shan-Shan, He Rui-Qing. Dynamic occlusion removal in single-pixel imaging system based on self-check. Acta Physica Sinica, 2023, 72(3): 034201. doi: 10.7498/aps.72.20221918
[7]	Hu Jin-Hu, Lin Dan-Ying, Zhang Wei, Zhang Chen-Shuang, Qu Jun-Le, Yu Bin. Dual-sided illumination light-sheet fluorescence microscopy with virtual single-pixel imaging deconvolution. Acta Physica Sinica, 2022, 71(2): 028701. doi: 10.7498/aps.71.20211358
[8]	Zhan Qing-Liang, Bai Chun-Jin, Ge Yao-Jun. Deep learning representation of flow time history for complex flow field. Acta Physica Sinica, 2022, 71(22): 224701. doi: 10.7498/aps.71.20221314
[9]	Cui An-Jing, Li Dao-Jing, Wu Jiang, Zhou Kai, Gao Jing-Han. Sparse sampling in frequency domain and laser imaging. Acta Physica Sinica, 2022, 71(5): 058705. doi: 10.7498/aps.71.20211408
[10]	Zhao Wei-Rui, Wang Hao, Zhang Lu, Zhao Yue-Jin, Chu Chun-Yan. High-precision co-phase method for segments based on a convolutional neural network. Acta Physica Sinica, 2022, 71(16): 164202. doi: 10.7498/aps.71.20220434
[11]	Sui Yi-Hui, Guo Xing-Yi, Yu Jun-Jin, Alexander A. Solovev, Ta De-An, Xu Kai-Liang. Accelerating super-resolution ultrasound localization microscopy using generative adversarial net. Acta Physica Sinica, 2022, 71(22): 224301. doi: 10.7498/aps.71.20220954
[12]	Zhou Jing, Zhang Xiao-Fang, Zhao Yan-Geng. Phase retrieval wavefront sensing based on image fusion and convolutional neural network. Acta Physica Sinica, 2021, 70(5): 054201. doi: 10.7498/aps.70.20201362
[13]	Huang Wei-Jian, Li Yong-Tao, Huang Yuan. Prediction of chaotic time series using hybrid neural network and attention mechanism. Acta Physica Sinica, 2021, 70(1): 010501. doi: 10.7498/aps.70.20200899
[14]	Wang Chen-Yang, Duan Qian-Qian, Zhou Kai, Yao Jing, Su Min, Fu Yi-Chao, Ji Jun-Yang, Hong Xin, Liu Xue-Qin, Wang Zhi-Yong. A hybrid model for photovoltaic power prediction of both convolutional and long short-term memory neural networks optimized by genetic algorithm. Acta Physica Sinica, 2020, 69(10): 100701. doi: 10.7498/aps.69.20191935
[15]	Li Ming-Fei, Yan Lu, Yang Ran, Liu Yuan-Xing. Fast single-pixel imaging based on optimized reordering Hadamard basis. Acta Physica Sinica, 2019, 68(6): 064202. doi: 10.7498/aps.68.20181886
[16]	Zhou Yang, Zhang Hong-Wei, Zhong Fei, Guo Shu-Xu. Iterative denoising of ghost imaging based on adaptive threshold method. Acta Physica Sinica, 2018, 67(24): 244201. doi: 10.7498/aps.67.20181240
[17]	Yao Wei-Qiang, Huang Wen-Hao, Yang Chu-Ping. Theoretical analysis of spectrum reconstruction imaging using single-pixel detection. Acta Physica Sinica, 2017, 66(3): 034201. doi: 10.7498/aps.66.034201
[18]	Liu Xin, Yi Ming-Hao, Guo Jin-Chuan. Line focal X-ray source imaging. Acta Physica Sinica, 2016, 65(21): 219501. doi: 10.7498/aps.65.219501
[19]	Yu Shu-Hai, Dong Lei, Liu Xin-Yue, Ling Jian-Yong. Analysis on reconstruction of virtual images of Fourier telescopy. Acta Physica Sinica, 2015, 64(18): 184205. doi: 10.7498/aps.64.184205
[20]	Cao Bei, Luo Xiu-Juan, Si Qing-Dan, Zeng Zhi-Hong. Four-phase closure algorithm for coherent field imaging. Acta Physica Sinica, 2015, 64(5): 054204. doi: 10.7498/aps.64.054204

Catalog

Vol. 74, Issue 8 - 2025-04-20

Metrics

Abstract views: 764
PDF Downloads: 17
Cited By: 0

姓名
邮箱
手机号码
标题
留言内容
验证码

Search

Article

留言板

Abstract

Authors and contacts

Corresponding author: CHEN Xihao, xi-haochen@163.com

FullText HTML

References

Cited By

Catalog

Metrics

Authors

Referees

About Journal

About

Search

Article

留言板

Abstract

Authors and contacts

Corresponding author: CHEN Xihao, xi-haochen@163.com

FullText HTML

References

Cited By

Catalog

Metrics

Publishing process

Authors

Referees

About Journal

About