Multi-scale/multi-resolution Kronecker compressive imaging

Canh, Thuong Nguyen; Dinh, Khanh Quoc

doi:10.1109/icip.2015.7351293

Cited by 14 publications

(19 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…We have compared the proposed GSR-NCR against six other competing approaches including BCS [11], BM3D-CS [12], ADS-CS [17], SGSR [26], ALSB [20] and MRK [27]. Note that ADS-CS and ALSB are patch-based sparse representation methods for image CS reconstruction.…”

Section: Performance Comparison With the State-of-the-art Methodsmentioning

confidence: 99%

“…Specifically, to investigate the sensi- [11], BM3D-CS [12], ADS-CS [17], SGSR [26], ALSB [20], MRK [27] and the Proposed GSR-NCR. Table 3: FSIM Comparisons of BCS [11], BM3D-CS [12], ADS-CS [17], SGSR [26], ALSB [20], MRK [27] and the Proposed GSR-NCR. tivity of our method against c, two experiments were conducted with respect to different c, ranging from 20 to 160, in the case of 0.2N and 0.3N measurements, respectively.…”

Section: Effect Of the Number Of The Best Matched Patchesmentioning

confidence: 99%

See 1 more Smart Citation

Group-based sparse representation for image compressive sensing reconstruction with non-convex regularization

et al. 2018

View full text Add to dashboard Cite

Patch-based sparse representation modeling has shown great potential in image compressive sensing (CS) reconstruction. However, this model usually suffers from some limits, such as dictionary learning with great computational complexity, neglecting the relationship among similar patches. In this paper, a groupbased sparse representation method with non-convex regularization (GSR-NCR) for image CS reconstruction is proposed. In GSR-NCR, the local sparsity and nonlocal self-similarity of images is simultaneously considered in a unified framework. Different from the previous methods based on sparsity-promoting convex regularization, we extend the non-convex weighted p (0< p <1) penalty function on group sparse coefficients of the data matrix, rather than conventional 1 -based regularization. To reduce the computational complexity, instead of learning the dictionary with a high computational complexity from natural images, we learn the principle component analysis (PCA) based dictionary for each group. Moreover, to make the proposed scheme tractable and robust, we have developed an efficient iterative shrinkage/thresholding algorithm to solve the non-convex optimization problem. Experimental results demonstrate that the proposed method outperforms many state-of-the-art techniques for image CS reconstruction. $ Fully documented templates are available in the elsarticle package on CTAN. Compressive sensing (CS) [1, 2, 3], which aims to recover signals from fewer measurements than suggested by the Nyquist sampling theory, is based on the hypothesis that the signals in question have compressible representations. The most attractive aspect of CS-based compression is that the sampling and compression are conducted simultaneously, and almost all computational cost is derived from the decoder stage, and thus, leading to a low computational cost of the encoder stage. Due to the superior property of CS, it has been widely applied to various areas, such as MRI image [4], remoting sensing [5], single-pixel camera [6] and sensor networks [7].In the theory of CS, if a signal is sparse in some transform domain, it is often sampled by the random projection and reconstructed by solving the 0 minimization problem with the prior information which usually makes up the regularization terms. However, due to the fact that 0 minimization is a difficult combinatorial optimization problem, solving this problem is NP-hard. For this reason, it has been proposed to replace the 0 norm by its convex 1 counterpart to make the optimization easy. For instance, Candès et al. [1] proposed that solving 1 minimization problem can recover a K-sparse signal X ∈ N from M = O(Klog(N/K)) random measurements. To solve the above 1 minimization problem, many CS reconstruction algorithms have been proposed, such as linear programming [3], gradient projection sparse reconstruction [8], match pursuit [9] and iterative thresholding [10].As a basic image inverse problem in the filed of image restoration, maybe the hottest topic is image CS reconstruction, which has a...

show abstract

Section: Performance Comparison With the State-of-the-art Methodsmentioning

confidence: 99%

Section: Effect Of the Number Of The Best Matched Patchesmentioning

confidence: 99%

Group-based sparse representation for image compressive sensing reconstruction with non-convex regularization

et al. 2018

View full text Add to dashboard Cite

show abstract

“…Recent works [5][6][7][8] have proven that multi-scale CS can be the optimal sampling solution. Radial Fourier subsampling [1] can be considered as a simple example and often used in bioimaging due to its physically driven projection.…”

Section: A Multi-scale Compressive Sensingmentioning

confidence: 99%

“…This work uses DIV2K [16] for training with 64x500 patches of size 256x256 and implement with MatConvNet [18], tested with 6 test images of classic512, Set5, and Set14 (*) . For conventional CS, we use BCS with GSR [18], Kronecker CS with DETER [19], and multi-scale KCS with MRKCS [7]. For DL-based CS, we use single scale BCS as ReconNet [11], DR 2 Net [12], DBCS [13], and CSNet [14].…”

Section: A Simulation Settingmentioning

confidence: 99%

“…For example, CS cannot assume prior information on the to-be-sampled signal beyond general sparsity prior. Therefore, based on the very general observation of the human visual system, a low-frequency prior is typically assumed, which has motivated researchers to develop multiscale CS [5][6][7][8] which captures more low-frequency components in image/video signalsinput image signal is linearly decomposed into multi-scales and sampling is done adaptively to each scale.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Multi-Scale Deep Compressive Sensing Network

Canh

2018

2018 IEEE Visual Communications and Image Processing (VCIP)

Self Cite

View full text Add to dashboard Cite

With joint learning of sampling and recovery, the deep learning-based compressive sensing (DCS) has shown significant improvement in performance and running time reduction. Its reconstructed image, however, losses highfrequency content especially at low subrates. This happens similarly in the multi-scale sampling scheme which also samples more low-frequency components. In this paper, we propose a multi-scale DCS convolutional neural network (MS-DCSNet) in which we convert image signal using multiple scale-based wavelet transform, then capture it through convolution block by block across scales. The initial reconstructed image is directly recovered from multi-scale measurements. Multi-scale wavelet convolution is utilized to enhance the final reconstruction quality. The network is able to learn both multi-scale sampling and multi-scale reconstruction, thus results in better reconstruction quality.

show abstract