• Journal of Broadcast Engineering
• /
• v.21 no.1
• /
• pp.60-65
• /
• 2016

This paper introduces the non-local means (NLM) algorithm for image denoising, and also introduces an improved algorithm which is based on the principal component analysis (PCA). To do the PCA, a covariance matrix of a given image should be evaluated first. If we let the size of neighborhood patches of the NLM S × S², and let the number of pixels Q, a matrix multiplication of the size S² × Q is required to compute a covariance matrix. According to the characteristic of images, such computation is inefficient. Therefore, this paper proposes an efficient method to compute the covariance matrix by sampling the pixels. After sampling, the covariance matrix can be computed with matrices of the size S² × floor (Width/l) × (Height/l).

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.10
• /
• pp.1317-1322
• /
• 2021

Nonlocal similarity of natural images leads to the fact that a patch matrix whose columns are similar patches of the reference patch has a low rank. Images corrupted by additive white Gaussian noises (AWGN) make their patch matrices to have a higher rank. The noise in the image can be reduced by obtaining low rank approximation of the patch matrices. In this paper, an image denoising algorithm is proposed, which first constructs the patch matrices by combining the similar patches of each reference patch, which is a part of the noisy image. For each patch matrix, the proposed algorithm calculates its low rank approximate, and then recovers the original image by aggregating the low rank estimates. The simulation results using widely accepted test images show that the proposed denoising algorithm outperforms four recent methods.

• ETRI Journal
• /
• v.36 no.1
• /
• pp.167-170
• /
• 2014

In this letter, we propose an unsupervised framework for speech noise reduction based on the recent development of low-rank and sparse matrix decomposition. The proposed framework directly separates the speech signal from noisy speech by decomposing the noisy speech spectrogram into three submatrices: the noise structure matrix, the clean speech structure matrix, and the residual noise matrix. Evaluations on the Noisex-92 dataset show that the proposed method achieves a signal-to-distortion ratio approximately 2.48 dB and 3.23 dB higher than that of the robust principal component analysis method and the non-negative matrix factorization method, respectively, when the input SNR is -5 dB.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.6
• /
• pp.2634-2648
• /
• 2020

In this paper, an adaptive sparse singular value decomposition (ASSVD) algorithm is proposed to estimate the signal matrix when only one data matrix is observed and there is high dimensional white noise, in which we assume that the signal matrix is low-rank and has sparse singular vectors, i.e. it is a simultaneously low-rank and sparse matrix. It is a structured matrix since the non-zero entries are confined on some small blocks. The proposed algorithm estimates the singular values and vectors separable by exploring the structure of singular vectors, in which the recent developments in Random Matrix Theory known as anisotropic Marchenko-Pastur law are used. And then we prove that when the signal is strong in the sense that the signal to noise ratio is above some threshold, our estimator is consistent and outperforms over many state-of-the-art algorithms. Moreover, our estimator is adaptive to the data set and does not require the variance of the noise to be known or estimated. Numerical simulations indicate that ASSVD still works well when the signal matrix is not very sparse.

• The Journal of the Acoustical Society of Korea
• /
• v.38 no.5
• /
• pp.534-542
• /
• 2019

In active SONAR, several different methods are used to detect range-Doppler information of the target. Compressive sensing based method is more accurate than conventional methods and shows superior performance. There are several compressive sensing algorithms for range-Doppler estimation of active sonar. The ability of each algorithm depends on algorithm type, mutual coherence of sensing matrix, and signal to noise ratio. In this paper, we compared and analyzed computational performance and accuracy of various compressive sensing algorithms for range-Doppler estimation of active sonar. The performance of OMP (Orthogonal Matching Pursuit), CoSaMP (Compressive Sampling Matching Pursuit), BPDN (CVX) (Basis Pursuit Denoising), LARS (Least Angle Regression) algorithms is respectively estimated for varying SNR (Signal to Noise Ratio), and mutual coherence. The optimal compressive sensing algorithm is presented according to the situation.

• Bulletin of the Korean Chemical Society
• /
• v.24 no.9
• /
• pp.1345-1350
• /
• 2003

Principal component analysis based two-dimensional (PCA-2D) correlation analysis is applied to FTIR spectra of polystyrene/methyl ethyl ketone/toluene solution mixture during the solvent evaporation. Substantial amount of artificial noise were added to the experimental data to demonstrate the practical noise-suppressing benefit of PCA-2D technique. 2D correlation analysis of the reconstructed data matrix from PCA loading vectors and scores successfully extracted only the most important features of synchronicity and asynchronicity without interference from noise or insignificant minor components. 2D correlation spectra constructed with only one principal component yield strictly synchronous response with no discernible a asynchronous features, while those involving at least two or more principal components generated meaningful asynchronous 2D correlation spectra. Deliberate manipulation of the rank of the reconstructed data matrix, by choosing the appropriate number and type of PCs, yields potentially more refined 2D correlation spectra.

• Proceedings of the IEEK Conference
• /
• 2003.07e
• /
• pp.1711-1714
• /
• 2003

The noise in an image degrades image quality and deteriorates coding efficiency of compression. Recently, various edge-preserving noise filtering methods based on the nonstationary image model have been proposed to overcome this problem. In most conventional nonstationary image models, however, pixels are assumed to be uncorrelated to each other In order not to increase the computational burden too much. As a result, some detailed information is lost in the filtered results. In this paper, we propose a computationally feasible adaptive noise smoothing algorithm which considers the nonstationary correlation characteristics of images. We assume that an image has a nonstationary mean and can be segmented into subimages which have individually different stationary correlations. Taking advantage of the special structure of the covariance matrix that results from the proposed image model, we derive a computationally efficient FFT-based adaptive linear minimum mean square error filter. The justification for the proposed image model is presented and the effectiveness of the proposed algorithm is demonstrated experimentally.

• Journal of Communications and Networks
• /
• v.12 no.4
• /
• pp.289-307
• /
• 2010

The problem of model selection arises in a number of contexts, such as subset selection in linear regression, estimation of structures in graphical models, and signal denoising. This paper studies non-asymptotic model selection for the general case of arbitrary (random or deterministic) design matrices and arbitrary nonzero entries of the signal. In this regard, it generalizes the notion of incoherence in the existing literature on model selection and introduces two fundamental measures of coherence-termed as the worst-case coherence and the average coherence-among the columns of a design matrix. It utilizes these two measures of coherence to provide an in-depth analysis of a simple, model-order agnostic one-step thresholding (OST) algorithm for model selection and proves that OST is feasible for exact as well as partial model selection as long as the design matrix obeys an easily verifiable property, which is termed as the coherence property. One of the key insights offered by the ensuing analysis in this regard is that OST can successfully carry out model selection even when methods based on convex optimization such as the lasso fail due to the rank deficiency of the submatrices of the design matrix. In addition, the paper establishes that if the design matrix has reasonably small worst-case and average coherence then OST performs near-optimally when either (i) the energy of any nonzero entry of the signal is close to the average signal energy per nonzero entry or (ii) the signal-to-noise ratio in the measurement system is not too high. Finally, two other key contributions of the paper are that (i) it provides bounds on the average coherence of Gaussian matrices and Gabor frames, and (ii) it extends the results on model selection using OST to low-complexity, model-order agnostic recovery of sparse signals with arbitrary nonzero entries. In particular, this part of the analysis in the paper implies that an Alltop Gabor frame together with OST can successfully carry out model selection and recovery of sparse signals irrespective of the phases of the nonzero entries even if the number of nonzero entries scales almost linearly with the number of rows of the Alltop Gabor frame.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.14 no.5
• /
• pp.2393-2399
• /
• 2013

In this paper, we proposed a method to screening of Alzheimer's disease (AD) from Raman spectra of platelet with synthesis of basis spectra using singular value decomposition (SVD). Raman spectra of platelet from AD transgenic mice are preprocessed with denoising, removal background and normalization method. The column vectors of each data matrix consist of Raman spectrum of AD and normal (NR). The matrix is factorized using SVD algorithm and then the basis spectra of AD and NR are determined by 12 column vectors of each matrix. The classification process is completed by select the class that minimized the root-mean-square error between the validation spectrum and the linear synthesized spectrum of the basis spectra. According to the experiments involving 278 Raman spectra, the proposed method gave about 97.6% classification rate, which is better performance about 6.1% than multi-layer perceptron (MLP) with extracted features using principle components analysis (PCA). The results show that the basis spectra using SVD is well suited for the diagnosis of AD by Raman spectra from platelet.