Qingwei Gao

A novel image denoising algorithm using linear Bayesian MAP estimation based on sparse representation

A novel image denoising algorithm using linear Bayesian maximum a posteriori (MAP) estimation based on sparse representation model is proposed. Starting from constructing prior probability distribution in representation vector, a linear Bayesian MAP estimator is constructed in order to acquire the most probable one behind the observations, which is adaptive to solve the generalized image inverse problems. Furthermore, a practical closed-form solution by affording some plausible approximations is obtained, and thus image denoising as a specialization can be easily solved. With our new method, we first extract all possible patches from noisy images and classify them to several sub-groups by their structural patterns, then train a different dictionary per each using the K-SVD algorithm, following by estimating corresponding parameters in MAP estimator. The final denoised image is obtained by applying denoising on each sub-group based on the estimator and averaging these outputs together. Simulated results show that the proposed method achieves a very competitive performance both in subjective visual quality and objective PSNR value, compared with other state-of-the-art denoising algorithms.

Directionlet-based denoising of SAR images using a Cauchy model

A new denoising algorithm based on directionlet transform using a Cauchy probability density function (PDF) is proposed to remove speckle noise. First, an anisotropic directionlet transform is taken on the logarithmically transformed SAR images. The directionlet transform coefficients of reflectance image are modeled as a zero-location Cauchy PDF, while the distribution of speckle noise is modeled as an additive Gaussian distribution with zero-mean. Then a maximum a posteriori (MAP) estimator is designed using the assumed priori models. And a regression-based method is proposed to estimate the parameters from the noisy observations. Finally, the performance of the proposed algorithm is compared with those of existing despeckling methods applied on both synthetic speckled images and actual SAR images. Experimental results show that the proposed scheme efficiently removes speckle noise from SAR images. Graphical abstractDisplay Omitted Highlights? Directionlet transform based on lattice is used to represent the SAR image. ? Cauchy distribution with one parameter is employed to fit the detail coefficients. ? Parameter is estimated in frequency domain instead of spatial domain. ? Edge preservation and ratio image are used as evaluation indexes of despeckling.

A large scale test dataset to determine optimal retention index threshold based on three mass spectral similarity measures.

Retention index (RI) is useful for metabolite identification. However, when RI is integrated with mass spectral similarity for metabolite identification, many controversial RI threshold setup are reported in literatures. In this study, a large scale test dataset of 5844 compounds with both mass spectra and RI information were created from National Institute of Standards and Technology (NIST) repetitive mass spectra (MS) and RI library. Three MS similarity measures: NIST composite measure, the real part of Discrete Fourier Transform (DFT.R) and the detail of Discrete Wavelet Transform (DWT.D) were used to investigate the accuracy of compound identification using the test dataset. To imitate real identification experiments, NIST MS main library was employed as reference library and the test dataset was used as search data. Our study shows that the optimal RI thresholds are 22, 15, and 15 i.u. for the NIST composite, DFT.R and DWT.D measures, respectively, when the RI and mass spectral similarity are integrated for compound identification. Compared to the mass spectrum matching, using both RI and mass spectral matching can improve the identification accuracy by 1.7%, 3.5%, and 3.5% for the three mass spectral similarity measures, respectively. It is concluded that the improvement of RI matching for compound identification heavily depends on the method of MS spectral similarity measure and the accuracy of RI data.

A high quality single-image super-resolution algorithm based on linear Bayesian MAP estimation with sparsity prior

This paper proposes a novel single-image super-resolution algorithm based on linear Bayesian maximum a posteriori (MAP) estimation and sparse representation. Starting from constructing several probability distribution priors in representation vector, we develop a linear Bayesian MAP estimator to acquire the most probable high-resolution (HR) image behind the low-resolution (LR) observation. Our new algorithm involves three main steps: (1) obtaining an initial estimate of the HR image via bi-cubic interpolation algorithm, (2) performing sparse coding on the initial estimate to get the representation vector and its support, (3) using the MAP estimator to restore the desired representation vector and then reconstructing the HR output. Simulated results show that the proposed method can achieve a more competitive performance both in subjective visual quality and in peak-signal-to-noise ratio (PSNR) and structural similarity (SSIM) measures, compared with other state-of-the-art super-resolution methods.

Efficient video copy detection using multi-modality and dynamic path search

Abstract Efficient and robust video copy detection is an important topic for many applications, such as commercial monitoring and social media retrieval. In this paper, with the aim of handling large-scale video data, we propose an efficient and robust video copy detection method jointly utilizing the characteristics of temporal continuity and multi-modality of video. The video is converted to a continuous sequence of states, and both the visual and auditory features are extracted for temporal frames. To facilitate tolerance of the length variations caused during video re-targeting, an efficient dynamic path search method is proposed to detect the target video clips, and highly compact audio fingerprint and visual ordinal features are jointly utilized in a flexible frame. The proposed scheme not only achieves high computational efficiency but also guarantees effectiveness in real applications. Comparison experiments were conducted using video commercials and real television programs from four channels as well as a benchmark video copy detection dataset, and the results demonstrate both the high efficiency and high robustness of the proposed method.

Directionlet-based method using the Gaussian mixture prior to SAR image despeckling

In this article, a new denoising algorithm is proposed based on the directionlet transform and the maximum a posteriori (MAP) estimation. The detailed directionlet coefficients of the logarithmically transformed noise-free image are considered to be Gaussian mixture probability density functions (PDFs) with zero means, and the speckle noise in the directionlet domain is modelled as additive noise with a Gaussian distribution. Then, we develop a Bayesian MAP estimator using these assumed prior distributions. Because the estimator that is the solution of the MAP equation is a function of the parameters of the assumed mixture PDF models, the expectation-maximization (EM) algorithm is also utilized to estimate the parameters, including weight factors and variances. Finally, the noise-free SAR image is restored from the estimated coefficients yielded by the MAP estimator. Experimental results show that the directionlet-based MAP method can be successfully applied to images and real synthetic aperture radar images to denoise speckle.

Tumor Classification Using Eigengene-Based Classifier Committee Learning Algorithm

Eigengene extracted by independent component analysis (ICA) is one kind of effective feature for tumor classification. In this letter, a novel tumor classification approach is proposed by using eigengene and support vector machine (SVM) based classifier committee learning (CCL) algorithm. In this method, a strategy of random feature subspace division is designed to improve the diversity of weaker classifiers. Gene expression data constructed by different feature subspaces are modeled by ICA, respectively. And the corresponding eigengene sets extracted by the ICA algorithm are used as the inputs of the weaker SVM classifiers. Moreover, a strategy of Bayesian sum rule (BSR) is designed to integrate the outputs of the weaker SVM classifiers, and used to provide a final decision for the tumor category. Experimental results on three DNA microarray datasets demonstrate that the proposed method is effective and feasible for tumor classification.

SAR speckle reduction using Laplace mixture model and spatial mutual information in the directionlet domain

The reduction of multiplicative speckle noise which always complicates the human and automatic interpretation of objects is very significant for the practical applications of synthetic aperture radar (SAR) image. In this paper, a new maximum a posteriori (MAP) despeckling method based on directionlet transform is proposed. To convert the multiplicative noise into an additive one, the logarithmic transform is first applied to the SAR images. Then, the directionlet coefficients of the noise-free (or underlying backscatter) image and of the speckle noise are modeled as Laplace mixture distribution with zero-mean and Gaussian distribution, respectively. Within Bayesian framework, a MAP estimator is constructed using these assumed prior distributions. After obtaining the parameter estimates using expectation-maximization algorithm, the noise-free coefficients are estimated by a non-linear shrinkage function based on the average version of Bayesian estimator. To improve the denoising performance, we combine the intra-scale dependency in terms of mutual information with the MAP estimator to refine the estimated results. Finally, we compare the proposed algorithm with several other speckle filters applied on synthetic and actual SAR images. Experimental results show that the proposed method outperforms other filters in terms of signal-to-noise ratio, edge preservation and equivalent number of looks measures in most cases.

Image interpolation via collaging its non-local patches

Abstract Interpolation is an important problem in image processing. The main issue on this application is to recover high frequency components lost by aliasing. In this paper, a novel spatial interpolation method exploiting the non-local redundancy of image is proposed, where the high-resolution (HR) image can be reconstructed by collaging the patches of its low-resolution (LR) observation. The appeal in this work is its simplicity, with no requirement of solving complex optimization equations. Simulation results suggest that the proposed method achieves a very competitive performance in both subjective visual quality and objective image quality (in terms of PSNR and structural similarity index measurement (SSIM)), compared to some recently proposed structured sparse representation-based methods.

A denoising method based on null space pursuit for infrared spectrum

Abstract The reduction of noise in infrared spectrum is of vital importance, as the infrared signal is often buried in excessive noise. In this paper, a novel denoising method based on null space pursuit (NSP) is proposed to eliminate noise in the infrared signal. The NSP is an adaptive operator-based signal separation approach, which can decompose signal into subband components and residue according to their characteristics. The residue is processed by mean filter because it contains very little useful information. Then, the subband components and the filtered residue are used to reconstruct the noise-free signal. Experimental results show that the proposed denoising method is effective in suppressing noise while retaining signal characteristics.