Non-local Means Denoising Research Articles

Non-local mean denoising in diffusion tensor space.

The aim of the present study was to present a novel non-local mean (NLM) method to denoise diffusion tensor imaging (DTI) data in the tensor space. Compared with the original NLM method, which uses intensity similarity to weigh the voxel, the proposed method weighs the voxel using tensor similarity measures in the diffusion tensor space. Euclidean distance with rotational invariance, and Riemannian distance and Log-Euclidean distance with affine invariance were implemented to compare the geometric and orientation features of the diffusion tensor comprehensively. The accuracy and efficacy of the proposed novel NLM method using these three similarity measures in DTI space, along with unbiased novel NLM in diffusion-weighted image space, were compared quantitatively and qualitatively in the present study.

SU‐E‐QI‐05: Denoising Intravoxel Incoherent Motion Magnetic Resonance Images Using Non‐Local Mean Technique for Oropharyngeal Cancer Study

Purpose:Intravoxel incoherent motion (IVIM) magnetic resonance imaging (MRI) normally shows a low signal to noise ratio (SNR) due to the presence of noise which complicates and biases the estimation of quantitative diffusion parameters. In this study, a Non‐local Means (NLM) approach was applied to remove the noise in oropharyngeal cancer IVIMMRI images.Methods:Eight male patients with squamous cell carcinoma of the oropharynx were included in this study under an approved IRB protocol. IVIM‐MRI was carried out on a 3.0‐T GE MRI. NLM denoising technique was performed using the MIPAV (V7, NIH). IVIM parameters (D, pure diffusion coefficient; f, perfusion fraction; D*, pseudodiffusion coefficient) were calculated on a pixel‐by‐pixel basis using a bi‐exponential model implemented within ImageJ (V1.47, NIH) with ten b‐values ranging from 0 – 800 s/mm2. SNR with and without denoising processing was calculated and compared in tumor regions. The agreements of IVIM‐MRI parameters estimated between with and without NLM processing in tumor region were assessed by Bland‐Altman plots.Results:Examples of IVIM images (b=800 s/mm2) and parametric maps (D, D* and f) with and without NLM applied are illustrated in Fig. 1 for an oropharynx cancer patient. Results of SNR and IVIM diffusion parameters between two different processing, as mean and standard deviation on 8 patients, are reported in Table 1. Bland‐Altman plots (Fig. 2) between the two approaches show better concordance for D (0.22% ± 1.42%) and f (0.56% ± 3.49%), whereas larger discrepancies were found for D* (0.86% ± 6.25%).Conclusion:NLM approach gives better SNR and quantitative data quality, also shows a good agreement between the IVIM parameters processed with and without NLM approaches. Therefore, NLM denoising technique can be applied to improve performance in terms of denoising quality and estimation of IVIM parameter as a post‐processing step without increasing the scanning time.

Novel Speed-Up Strategies for Non-Local Means Denoising With Patch and Edge Patch Based Dictionaries

In this paper, a novel technique to speed-up a nonlocal means (NLM) filter is proposed. In the original NLM filter, most of its computational time is spent on finding distances for all the patches in the search window. Here, we build a dictionary in which patches with similar photometric structures are clustered together. Dictionary is built only once with high resolution images belonging to different scenes. Since the dictionary is well organized in terms of indexing its entries, it is used to search similar patches very quickly for efficient NLM denoising. We achieve a substantial reduction in computational cost compared with the original NLM method, especially when the search window of NLM is large, without much affecting the PSNR. Second, we show that by building a dictionary for edge patches as opposed to intensity patches, it is possible to reduce the dictionary size; thus, further improving the computational speed and memory requirement. The proposed method preclassifies similar patches with the same distance measure as used by NLM method. The proposed algorithm is shown to outperform other prefiltering based fast NLM algorithms computationally as well as qualitatively.

Non-Local Means Denoising of Dynamic PET Images

ObjectiveDynamic positron emission tomography (PET), which reveals information about both the spatial distribution and temporal kinetics of a radiotracer, enables quantitative interpretation of PET data. Model-based interpretation of dynamic PET images by means of parametric fitting, however, is often a challenging task due to high levels of noise, thus necessitating a denoising step. The objective of this paper is to develop and characterize a denoising framework for dynamic PET based on non-local means (NLM).TheoryNLM denoising computes weighted averages of voxel intensities assigning larger weights to voxels that are similar to a given voxel in terms of their local neighborhoods or patches. We introduce three key modifications to tailor the original NLM framework to dynamic PET. Firstly, we derive similarities from less noisy later time points in a typical PET acquisition to denoise the entire time series. Secondly, we use spatiotemporal patches for robust similarity computation. Finally, we use a spatially varying smoothing parameter based on a local variance approximation over each spatiotemporal patch.MethodsTo assess the performance of our denoising technique, we performed a realistic simulation on a dynamic digital phantom based on the Digimouse atlas. For experimental validation, we denoised PET images from a mouse study and a hepatocellular carcinoma patient study. We compared the performance of NLM denoising with four other denoising approaches – Gaussian filtering, PCA, HYPR, and conventional NLM based on spatial patches.ResultsThe simulation study revealed significant improvement in bias-variance performance achieved using our NLM technique relative to all the other methods. The experimental data analysis revealed that our technique leads to clear improvement in contrast-to-noise ratio in Patlak parametric images generated from denoised preclinical and clinical dynamic images, indicating its ability to preserve image contrast and high intensity details while lowering the background noise variance.

Nonlocal denoising using anisotropic structure tensor for 3D MRI

Noise in magnetic resonance imaging (MRI) data is widely recognized to be harmful to image processing and subsequent quantitative analysis. To ameliorate the effects of image noise, the authors present a structure-tensor based nonlocal mean (NLM) denoising technique that can effectively reduce noise in MRI data and improve tissue characterization. The proposed 3D NLM algorithm uses a structure tensor to characterize information around tissue boundaries. The similarity weight of a pixel (or patch), which determines its contribution to the denoising process, is determined by the intensity and structure tensor simultaneously. Meanwhile, similarity of structure tensors is computed using an affine-invariant Riemannian metrics, which compares tensor properties more comprehensively and avoids orientation inaccuracy of structure subsequently. The proposed method is further extended for denoising high dimensional MRI data such as diffusion weighted MRI. It is also extended to handle Rician noise corruption so that denoising effects are further enhanced. The proposed method was implemented in both simulated datasets and multiply modalities of real 3D MRI datasets. Comparisons with related state-of-the-art algorithms demonstrated that this method improves denoising performance qualitatively and quantitatively. In this paper, high order structure information of 3D MRI was characterized by 3D structure tensor and compared for NLM denoising in a Riemannian space. Experiments with simulated and real human MRI data demonstrate a great potential of the proposed technique for routine clinical use.

Denoising and fast diffusion imaging with physically constrained sparse dictionary learning

Diffusion-weighted imaging (DWI) allows imaging the geometry of water diffusion in biological tissues. However, DW images are noisy at high b-values and acquisitions are slow when using a large number of measurements, such as in Diffusion Spectrum Imaging (DSI). This work aims to denoise DWI and reduce the number of required measurements, while maintaining data quality. To capture the structure of DWI data, we use sparse dictionary learning constrained by the physical properties of the signal: symmetry and positivity. The method learns a dictionary of diffusion profiles on all the DW images at the same time and then scales to full brain data. Its performance is investigated with simulations and two real DSI datasets. We obtain better signal estimates from noisy measurements than by applying mirror symmetry through the q-space origin, Gaussian denoising or state-of-the-art non-local means denoising. Using a high-resolution dictionary learnt on another subject, we show that we can reduce the number of images acquired while still generating high resolution DSI data. Using dictionary learning, one can denoise DW images effectively and perform faster acquisitions. Higher b-value acquisitions and DSI techniques are possible with approximately 40 measurements. This opens important perspectives for the connectomics community using DSI.

An Improvement of Non-Local Means Denoising Method in the Presence of Large Noise

The performance of singular value decomposition (SVD) based nonlocal mean (NLM) denoising method degrades when the noise is high. This paper describes an approach of an improvement of NLM denoising when the noise is large. Instead of SVD, we combine the kernel principal component analysis (KPCA) with NLM. It is demonstrated in terms of peak signal to noise ratio (PSNR) in decibels (dB) that the NLM denoising method is improved using various test images corrupted by large additive white Gaussian noise (AWGN)

A New Weight for Nonlocal Means Denoising Using Method Noise

Method noise, i.e., the difference between the noisy image and its denoised version, often contains residual image information due to imperfect denoising methods. In this letter, a novel weight is derived for nonlocal means (NLM) denoising, which tries to exploit the nonlocal similarities of residual image in method noise, and estimates directly the similarities between noise free patches instead of the commonly used similarity measure based on noisy observations or their pre-denoised versions. The denoising scheme is implemented in two stages: the first stage is the original NLM, and the second stage is the NLM with the new weight that incorporates the role of both the pre-denoised result and the method noise produced in the previous stage. Experimental results demonstrate the robustness of the proposed weight and its potential with respect to classical NLM methods and other state-of-the-art methods.

Rotationally invariant similarity measures for nonlocal image denoising

Many natural or texture images contain structures that appear several times in the image. One of the denoising filters that successfully take advantage of such repetitive regions is NL means. Unfortunately, the block matching of NL means cannot handle rotation or mirroring. In this paper, we analyse two natural approaches for a rotationally invariant similarity measure that will be used as an alternative to, respectively a modification of the well-known block matching algorithm in nonlocal means denoising. The first approach is based on moment invariants whereas the second one estimates the rotation angle, rotates the block via interpolation and then uses a standard block matching. In contrast to the standard method, the presented algorithms can find similar regions or patches in an image even if they appear in several rotated or mirrored instances. Hence, one can find more suitable regions for the weighted average and yield improved results.

An Optimized Blockwise Nonlocal Means Denoising Filter for 3-D Magnetic Resonance Images

A critical issue in image restoration is the problem of noise removal while keeping the integrity of relevant image information. Denoising is a crucial step to increase image quality and to improve the performance of all the tasks needed for quantitative imaging analysis. The method proposed in this paper is based on a 3-D optimized blockwise version of the nonlocal (NL)-means filter (Buades, et al., 2005). The NL-means filter uses the redundancy of information in the image under study to remove the noise. The performance of the NL-means filter has been already demonstrated for 2-D images, but reducing the computational burden is a critical aspect to extend the method to 3-D images. To overcome this problem, we propose improvements to reduce the computational complexity. These different improvements allow to drastically divide the computational time while preserving the performances of the NL-means filter. A fully automated and optimized version of the NL-means filter is then presented. Our contributions to the NL-means filter are: 1) an automatic tuning of the smoothing parameter; 2) a selection of the most relevant voxels; 3) a blockwise implementation; and 4) a parallelized computation. Quantitative validation was carried out on synthetic datasets generated with BrainWeb (Collins, et al., 1998). The results show that our optimized NL-means filter outperforms the classical implementation of the NL-means filter, as well as two other classical denoising methods [anisotropic diffusion (Perona and Malik, 1990)] and total variation minimization process (Rudin, et al., 1992) in terms of accuracy (measured by the peak signal-to-noise ratio) with low computation time. Finally, qualitative results on real data are presented .