- 标题
- 摘要
- 关键词
- 实验方案
- 产品
-
Iterative Adaptive Nonconvex Low-Rank Tensor Approximation to Image Restoration Based on ADMM
摘要： In this paper, in order to recover more ?ner details of the image and to avoid the loss of image structure information for image restoration problem, we develop an iterative adaptive weighted core tensor thresholding (IAWCTT) approach based on the alternating direction method of multipliers (ADMM). By observing the decoupling property of the ADMM algorithm, we ?rst propose that the key step to image restoration is to tackle the denoising subproblem ef?ciently using appropriate prior information. Secondly, by analyzing the properties of the core tensor, we propose that low-rank tensor approximation can be implemented by penalizing the core tensor itself, instead of penalizing the CP rank, Tucker rank or the multilinear rank and Tubal rank. The IAWCTT approach is proposed to solve the denoising subproblem in the ADMM framework, and we claim that such an adaptive weighted scheme is equivalent to a kind of nonconvex penalty for the core tensor; thus, it is unnecessary to use the nonconvex penalty term to induce strong sparse/low-rank solution in image restoration optimization problem, because the scheme that selecting appropriate weights to the convex penalty term can also lead to strong sparse/low-rank solution. Numerical experiments show that our proposed model and algorithm are comparable to other state-of-the-art models and methods.
关键词： Image restoration,Low-rank tensor approximation,ADMM,Nonconvex penalty
更新于2025-09-23 15:23:52
-
Hyperspectral Mixed Denoising via Spectral Difference-Induced Total Variation and Low-Rank Approximation
摘要： Exploration of multiple priors on observed signals has been demonstrated to be one of the effective ways for recovering underlying signals. In this paper, a new spectral difference-induced total variation and low-rank approximation (termed SDTVLA) method is proposed for hyperspectral mixed denoising. Spectral difference transform, which projects data into spectral difference space (SDS), has been proven to be powerful at changing the structures of noises (especially for sparse noise with a specific pattern, e.g., stripes or dead lines present at the same position in a series of bands) in an original hyperspectral image (HSI), thus allowing low-rank techniques to get rid of mixed noises more efficiently without treating them as low-rank features. In addition, because the neighboring pixels are highly correlated and the spectra of homogeneous objects in a hyperspectral scene are always in the same low-dimensional manifold, we are inspired to combine total variation and the nuclear norm to simultaneously exploit the local piecewise smoothness and global low rankness in SDS for mixed noise reduction of HSI. Finally, the alternating direction methods of multipliers (ADMM) is employed to effectively solve the SDTVLA model. Extensive experiments on three simulated and two real HSI datasets demonstrate that, in terms of quantitative metrics (i.e., the mean peak signal-to-noise ratio (MPSNR), the mean structural similarity index (MSSIM) and the mean spectral angle (MSA)), the proposed SDTVLA method is, on average, 1.5 dB higher MPSNR values than the competitive methods as well as performing better in terms of visual effect.
关键词： ADMM,total variation,hyperspectral mixed denoising,low-rank approximation,spectral difference space
更新于2025-09-23 15:23:52
-
[IEEE 2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop (SAM) - Sheffield (2018.7.8-2018.7.11)] 2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop (SAM) - Restoration of Multilayered Single-Photon 3D Lidar Images
摘要： This paper proposes a new algorithm to restore 3D single-photon Lidar images obtained under challenging realistic scenarios which include imaging multilayered targets such as semi-transparent surfaces or imaging through obscurants such as scattering media (e.g., water, fog). The Data restoration and exploitation is achieved by minimising an appropriate cost-function accounting for the data Poisson statistics and the available prior knowledge regarding the depth and reflectivity estimates. The proposed algorithm takes into account (i) the non-local spatial correlations between pixels, by using a convex non-local total variation (TV) regularizer, and (ii) the clustered nature of the returned photons, by using a collaborative sparse prior. The resulting minimization problem is solved using the alternating direction method of multipliers (ADMM) that offers good convergence properties. The algorithm is validated using both synthetic and real data which show the benefit of the proposed strategy in the sparse regime due to a fast acquisition or in presence of a high background due to obscurants.
关键词： Poisson statistics,collaborative sparsity,ADMM,image restoration,NR3D,Lidar waveform
更新于2025-09-23 15:22:29
-
[IEEE 2018 IEEE 13th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP) - Aristi Village, Zagorochoria, Greece (2018.6.10-2018.6.12)] 2018 IEEE 13th Image, Video, and Multidimensional Signal Processing Workshop (IVMSP) - A Fast Parallel Algorithm for Convolutional Sparse Coding
摘要： The current leading algorithms for convolutional sparse coding are not inherently parallelizable, and therefore are not able to fully exploit modern multi-core architectures. We address this deficiency by developing a new algorithm that partitions the dictionary and the corresponding coefficient maps into groups, solving the main subproblems for all of the groups in parallel. Theoretical complexities and implementational details are discussed and validated with computational experiments, which indicate speed improvements by about a factor of 5, depending on the specific problem.
关键词： Convolutional Sparse Representations,ADMM,Convolutional Sparse Coding
更新于2025-09-23 15:22:29
-
[IEEE 2018 10th International Conference on Wireless Communications and Signal Processing (WCSP) - Hangzhou (2018.10.18-2018.10.20)] 2018 10th International Conference on Wireless Communications and Signal Processing (WCSP) - Hyperspectral Image Classification VIA a Joint Sparsity and Spatial Correlation Model
摘要： In this paper, a novel constrained Sparse Representation (SR) algorithm based on the joint sparsity and spatial correlation for hyperspectral image (HSI) classification is proposed. The coefficients in the sparse vector associated with the training samples in the structured dictionary exhibit the group sparsity continuity. However, this joint sparsity of the coefficient vector is not considered in the classical SR classifiers. In addition, spatial correlation has positive effect on HSI classification processing. Thus in the proposed SR model, we consider a joint sparsity regularization term to promote the joint sparsity of the sparse vectors and use space regularization to restrict spatial correlation of the output. The formulated problem is solved via the alternating direction method of multipliers (ADMM). Simulation results show that the proposed algorithm has the improved performance.
关键词： sparse representation,classification,Hyperspectral imagery,joint sparsity,ADMM
更新于2025-09-23 15:22:29
-
An Efficient and Fast Quantum State Estimator With Sparse Disturbance
摘要： A pure or nearly pure quantum state can be described as a low-rank density matrix, which is a positive semidefinite and unit-trace Hermitian. We consider the problem of recovering such a low-rank density matrix contaminated by sparse components, from a small set of linear measurements. This quantum state estimation task can be formulated as a robust principal component analysis (RPCA) problem subject to positive semidefinite and unit-trace Hermitian constraints. We propose an efficient and fast inexact alternating direction method of multipliers (I-ADMM), in which the subproblems are solved inexactly and hence have closed-form solutions. We prove global convergence of the proposed I-ADMM, and the theoretical result provides a guideline for parameter setting. Numerical experiments show that the proposed I-ADMM can recover state density matrices of 5 qubits on a laptop in 0.69 s, with 6 × 10^{-4} accuracy (99.38% fidelity) using 30% compressive sensing measurements, which outperforms existing algorithms.
关键词： quantum state estimation (QSE),Alternating direction method of multipliers (ADMM),robust principal component analysis (RPCA)
更新于2025-09-23 15:22:29
-
An Outlier-insensitive Unmixing Algorithm with Spatially Varying Hyperspectral Signatures
摘要： Effective hyperspectral unmixing (HU) is essential to the estimation of the underlying materials’ signatures (endmember signatures) and their spatial distributions (abundance maps) from a given image (data) of a hyperspectral scene. Recently, investigating HU under the non-negligible endmember variability (EV) and outlier effects (OE) has drawn extensive attention. Some state-of-the-art works either consider EV or consider OE, but none of them considers both EV and OE simultaneously. In this paper, we propose a novel HU algorithm, referred to as the variability/outlier-insensitive multi-convex unmixing (VOIMU) algorithm, that is robust against both EV and OE. Considering two suitable regularizers, a nonconvex minimization problem is formulated for which the perturbed linear mixing model (PLMM) proposed by Thouvenin et al., is used for modeling EV, while OE is implicitly handled by applying a p quasi-norm to the data fitting with 0 < p < 1. Then we reformulate it into a multi-convex problem which is then solved by the block coordinate decent (BCD) method, with convergence guarantee by casting it into the block successive upper bound minimization (BSUM) framework. The proposed VOIMU algorithm can yield a stationary-point solution with convergence guarantee, together with some intriguing information of potential outlier pixels though outliers are neither physically modeled in the above problem nor detected in the algorithm operation. Finally, we provide some simulation results and experimental results using real data to demonstrate the efficacy and practical applicability of the proposed VOIMU algorithm.
关键词： block successive upper bound minimization (BSUM),endmember variability,alternating direction method of multipliers (ADMM),outlier effects,block coordinate decent (BCD) method,Hyperspectral imaging
更新于2025-09-23 15:22:29
-
ADMM-CSNet: A Deep Learning Approach for Image Compressive Sensing
摘要： Compressive sensing (CS) is an effective technique for reconstructing image from a small amount of sampled data. It has been widely applied in medical imaging, remote sensing, image compression, etc. In this paper, we propose two versions of a novel deep learning architecture, dubbed as ADMM-CSNet, by combining the traditional model-based CS method and data-driven deep learning method for image reconstruction from sparsely sampled measurements. We ?rst consider a generalized CS model for image reconstruction with undetermined regularizations in undetermined transform domains, and then two ef?cient solvers using Alternating Direction Method of Multipliers (ADMM) algorithm for optimizing the model are proposed. We further unroll and generalize the ADMM algorithm to be two deep architectures, in which all parameters of the CS model and the ADMM algorithm are discriminatively learned by end-to-end training. For both applications of fast CS complex-valued MR imaging and CS imaging of real-valued natural images, the proposed ADMM-CSNet achieved favorable reconstruction accuracy in fast computational speed compared with the traditional and the other deep learning methods.
关键词： Compressive sensing,MR Imaging,deep learning,ADMM-CSNet,ADMM
更新于2025-09-23 15:21:01
-
Sectioning-based ADMM Imaging for Fast Node Communication with a Compressive Antenna
摘要： In this paper, a novel norm-1 regularized imaging technique, based on the Alternating Direction Method of Multipliers (ADMM) algorithm, is presented. This technique divides the sensing matrix of the imaging system in submatrices by columns. This method is capable of imaging metallic targets with a reduced amount of measurements, by sectioning the imaging domain into several regions and optimizing them individually in different nodes. This technique reduces the amount of information to be shared among the nodes, compared to the consensus ADMM that divides the sensing matrix into submatrices by rows. This makes the communication among the nodes faster for large imaging domains that need to be distributed into several computational nodes of a cluster. A Compressive Reflector Antenna (CRA) has been recently proposed as a low cost hardware to provide high sensing capacity. The use of the column-wise division ADMM algorithm in combination with the imaging capabilities of the CRA allows a distributed and quasi real-time imaging, while reducing the communications among the computational nodes.
关键词： Compressive antenna,nodes communication,distributed ADMM,real-time imaging,norm-1 regularization
更新于2025-09-19 17:15:36
-
Multi‐Resonance Induced Thermally Activated Delayed Fluorophores for Narrowband Green OLEDs
摘要： With the increment of data scale, distributed machine learning has received more and more attention. However, as the data grows, the dimension of the dataset will increase rapidly, which leads to the increment of the communication traffic in the distributed computing cluster and decreases the performance of the distributed algorithms. This paper proposes a message filtering strategy based on asynchronous alternating direction method of multipliers (ADMM), which can effectively reduce the communication time of the algorithm while ensuring the convergence of the algorithm. In this paper, a soft threshold filtering strategy based on L1 regularization is proposed to filter the parameter of master node, and a gradient truncation filtering strategy is proposed to filter the parameter of slave node. Besides, we update the algorithm asynchronously to reduce the waiting time of the master node. Experiments on large-scale sparse data show that our algorithm can effectively reduce the traffic of messages and make the algorithm reach convergence in a shorter time.
关键词： gradient truncation,distributed machine learning,message filtering,asynchronous update,ADMM,L1 regularization
更新于2025-09-19 17:13:59