Optimization Methods of Compressively Sensed Image Reconstruction Based on Single-Pixel Imaging

Wei, Ziran; Zhang, Jianlin; Xu, Zhihai; Liu, Yong

doi:10.3390/app10093288

Cited by 10 publications

(10 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, the conventional reconstruction strategies are low in efficiency in terms of computation time. Wei et al [16] suggested a new mathematical model that initializes the basis differently to create binary measurement patterns that are more optimised, which results in better performance. Furthermore, post-processing of the reconstructed image has been researched in [17].…”

Section: Background a Single Pixel Imaging (Spi)mentioning

confidence: 99%

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

2023

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

View full text Add to dashboard Cite

show abstract

“…However, the recently researched results show that optimized design of the measurement matrix based on minimizing the mutual coherence is not robust when the sparse representation error (SRE) is considered [18,28]. In [9,48,50], an optimized design method of measurement matrix is proposed according to the theory of minimum Frobenius norm.…”

Section: Related Workmentioning

confidence: 99%

Image parallel block compressive sensing scheme using DFT measurement matrix

Wang

Jiang

Chen

2022

Multimed Tools Appl

View full text Add to dashboard Cite

Compressive sensing (CS)-based image coding has been widely studied in the field of image processing. However, the CS-based image encoder has a significant gap in image reconstruction performance compared with the conventional image compression methods. In order to improve the reconstruction quality of CS-based image encoder, we proposed an image parallel block compressive sensing (BCS) coding scheme, which is based on discrete Cosine transform (DCT) sparse basis matrix and partial discrete Fourier transform (DFT) measurement matrix. In the proposed parallel BCS scheme, each column of an image block is sampled by the same DFT measurement matrix. Due to the complex property of DFT measurement matrix, the compressed image data is complex. Then, the real part and imaginary part of the resulting BCS data are quantized and transformed into two bit streams, respectively. At the reconstruction stage, the resulting two bit streams are transformed back into two real signals using inverse quantization operation. The resulting two real signals are combined into one complex signal, which is served as the input data of the CS reconstructed algorithm. The theoretical analysis based on minimum Frobenius norm method demonstrates that the proposed DFT measurement matrix outperforms the other conventional measurement matrices. The simulation results show that the reconstructed performance of the proposed DFT measurement matrix is better than that of the other conventional measurement matrices for the proposed parallel BCS. Specifically, we analyzed the impact of quantization on the reconstruction performance of CS. The experiment results show that the effect of the quantization on reconstruction performance in BCS framework can nearly be ignored.

show abstract

“…The usual choices to obtain a sparse representation of natural images are the wavelet transform or the DCT [ 47 , 48 ]. The former bases are known to be mutually incoherent with structured projection matrices or even random projections for the sensing step [ 36 , 37 , 49 ]. The DCT is a transform widely employed in image compression and also in the context of CS [ 50 ].…”

Section: Cs Formulationmentioning

confidence: 99%

Coded Aperture Hyperspectral Image Reconstruction

García-Sánchez

Fresnedo

González-Coma

et al. 2021

Sensors

View full text Add to dashboard Cite

In this work, we study and analyze the reconstruction of hyperspectral images that are sampled with a CASSI device. The sensing procedure was modeled with the help of the CS theory, which enabled efficient mechanisms for the reconstruction of the hyperspectral images from their compressive measurements. In particular, we considered and compared four different type of estimation algorithms: OMP, GPSR, LASSO, and IST. Furthermore, the large dimensions of hyperspectral images required the implementation of a practical block CASSI model to reconstruct the images with an acceptable delay and affordable computational cost. In order to consider the particularities of the block model and the dispersive effects in the CASSI-like sensing procedure, the problem was reformulated, as well as the construction of the variables involved. For this practical CASSI setup, we evaluated the performance of the overall system by considering the aforementioned algorithms and the different factors that impacted the reconstruction procedure. Finally, the obtained results were analyzed and discussed from a practical perspective.

show abstract

Optimization Methods of Compressively Sensed Image Reconstruction Based on Single-Pixel Imaging

Cited by 10 publications

References 57 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

Image parallel block compressive sensing scheme using DFT measurement matrix

Coded Aperture Hyperspectral Image Reconstruction

Contact Info

Product

Resources

About