Efficient Spatially-Variant Single-Pixel Imaging Using Block-Based Compressed Sensing

Shin, Zhenyong; Chai, Tong-Yuen; Pua, Chang Hong; Xin, Wang; Chua, Sing Yee

doi:10.1007/s11265-021-01689-5

Cited by 8 publications

(5 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides, an iterative total error compensation method [18] has been proposed to obtain high quality images. Block-based CS was studied for the reduction of computation cost [19]. More recently, as deep learning is becoming the hotbed for imaging-related problems, deep learning based SPI has been investigated [20]- [22].…”

Section: Background a Single Pixel Imaging (Spi)mentioning

confidence: 99%

Adaptive Coarse-to-Fine Single Pixel Imaging With Generative Adversarial Network Based Reconstruction

2023

Self Cite

View full text Add to dashboard Cite

Single Pixel Imaging (SPI) that only uses one light intensity sensor has been researched extensively as an alternative imaging method. It is proven to work at low light conditions and unconventional wavelength bands where the classical pixel array sensors are limited. However, the major issues of SPI remain as the image quality and computational efficiency. Thus, this paper proposes an adaptive coarse-to-fine (C2F) sampling method to replace the typical uniform sampling method to achieve image reconstruction with better quality. This scalable sampling mechanism is adaptive to the target scene as it will progressively sample according to the image complexity and quality indicator. Subsequently, a deep Generative Adversarial Network (GAN) model is also proposed to improve the time efficiency of the multiscale image reconstruction. The results show that C2F sampling consistently outperforms uniform sampling in terms of image quality (21% in SSIM, 8% in PSNR and 17% in RMSE). Besides, improvement in the efficiency is also achieved by the proposed GAN reconstruction, whereby the total time taken is only 0.025% of the time taken for the conventional L1 reconstruction method (≈ 4000 times faster). In conclusion, the proposed adaptive C2F SPI using GAN reconstruction method can serve as an optimised solution to improve both the image quality and computational efficiency in SPI.

show abstract

Section: Background a Single Pixel Imaging (Spi)mentioning

confidence: 99%

Adaptive Coarse-to-Fine Single Pixel Imaging With Generative Adversarial Network Based Reconstruction

2023

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, the problem can be mitigated by both hardware or software aspects. 9,10 Recently, by increasing the reference speed of the light modulator, the efficiency of SPI has been improved. Salvador-Balaguer et al 11 introduced a low-cost SPI imaging solution by using light-structured illumination generated with a high refresh rate low-cost color LED array controlled by a field programmable gate array system.…”

Section: Introductionmentioning

confidence: 99%

“…The acquisition of a large number of measurements is often necessary, resulting in low sampling efficiency. However, the problem can be mitigated by both hardware or software aspects 9 , 10 . Recently, by increasing the reference speed of the light modulator, the efficiency of SPI has been improved.…”

Section: Introductionmentioning

confidence: 99%

Self-adaptive sampling for Fourier single-pixel imaging using probability estimation

Tey,

Tham,

Phua

et al. 2024

J. Electron. Imag.

Self Cite

View full text Add to dashboard Cite

Fourier single-pixel imaging (FSI) produces images through the acquisition of Fourier domain information using a single-pixel detector and a sequence of light modulation patterns. Conventional sampling approach in FSI tends to perform poorly when it comes to capturing the intricate details present in high-frequency components of the image. The variable density sampling method follows a predefined mechanism where the power of image information decreases when frequency increases. To enhance the sampling efficiency, we propose a self-adaptive sampling method to dynamically determine the order of the illumination pattern based on the scene's spectrum distribution through the probability estimation of the low-frequency samples. The image is subsequently reconstructed through compressed sensing technique. Our results indicate improved image quality even at low sampling ratios. Unlike existing adaptive approaches, the proposed method does not require dataset training or redundant sampling. Instead, it exhibits remarkable versatility by adapting to various types of images.

show abstract

“…In recent years, compressed sensing (CS) has been successfully used in many imaging systems to significantly reduce the amount of data sampling, speed up signal acquisition, and alleviate the pressure of data processing. For instance, CS has enabled single-pixel imaging (SPI) [21][22][23], compressed ultrafast photography (CUP) [24][25][26], and compressive hyperspectral Raman imaging [27][28][29]. The success of these CS-based imaging methods relies on the fact that the signal usually contains only a few nonzero values in a particular domain (e.g., Fourier domain [30], discrete cosine domain [31], discrete wavelet domain [32]), and a small number of measurements are sufficient to capture them.…”

Section: Introductionmentioning

confidence: 99%

Optical time‐stretch imaging flow cytometry in the compressed domain

Lin

Weng

et al. 2023

Journal of Biophotonics

View full text Add to dashboard Cite

Imaging flow cytometry based on optical time‐stretch (OTS) imaging combined with a microfluidic chip attracts much attention in the large‐scale single‐cell analysis due to its high throughput, high precision, and label‐free operation. Compressive sensing has been integrated into OTS imaging to relieve the pressure on the sampling and transmission of massive data. However, image decompression brings an extra overhead of computing power to the system, but does not generate additional information. In this work, we propose and demonstrate OTS imaging flow cytometry in the compressed domain. Specifically, we constructed a machine‐learning network to analyze the cells without decompressing the images. The results show that our system enables high‐quality imaging and high‐accurate cell classification with an accuracy of over 99% at a compression ratio of 10%. This work provides a viable solution to the big data problem in OTS imaging flow cytometry, boosting its application in practice.

show abstract

Efficient Spatially-Variant Single-Pixel Imaging Using Block-Based Compressed Sensing

Cited by 8 publications

References 50 publications

Adaptive Coarse-to-Fine Single Pixel Imaging With Generative Adversarial Network Based Reconstruction

Adaptive Coarse-to-Fine Single Pixel Imaging With Generative Adversarial Network Based Reconstruction

Self-adaptive sampling for Fourier single-pixel imaging using probability estimation

Optical time‐stretch imaging flow cytometry in the compressed domain

Contact Info

Product

Resources

About