Fast Nonlocal Means Filtering Research Articles

Image denoising using fast non-local means filter and multi-thresholding with harmony search algorithm for WSN

Several image denoising algorithms are developed so far to obtain a better denoised sensor images. But they fail to preserve the edges of the images. In order to overcome this, an attempt has been made in this paper by incorporating the various filters like fast non-local means filter (FNLMF) and high boost filter with the existing wavelet thresholding-based denoising method. However, the denoised output and the computation time are affected by the wavelet properties significantly. Hence, this paper concentrates on histogram-based multi-thresholding (HMT) as a major part to denoise the image. Here, the corrupted image is first denoised by applying FNLMF filtering section. Then the edges and minimal details of the denoised output are enhanced by utilising the HMT with harmony search algorithm (HSA)-based optimisation technique. From the simulation results, the proposed denoising method can be adoptable for WSN.

Image denoising using fast non-local means filter and multi-thresholding with harmony search algorithm for WSN

Image denoising using fast non-local means filter and multi-thresholding with harmony search algorithm for WSN

An improved wavelet denoising algorithm for SAR interferogram using fast non-local means filtering

Due to the existence of SAR interferogram noise, the quality of the interferogram is seriously affected. Therefore, to suppress the influence of phase noise, many scholars have carried out a lot of research on SAR interferogram noise reduction, but existing studies have difficulty maintaining edge details while also considering effective noise reduction. Therefore, we proposed an improved wavelet denoising algorithm for SAR interferogram using fast non-local means filtering. The algorithm was divided into three steps. First, the appropriate wavelet basis was selected and the decomposition level was determined by white noise detection to achieve wavelet decomposition. Then, after non-local processing of wavelet coefficients, the threshold denoising rule was constructed by improving the threshold function to achieve the first denoising. Finally, fast non-local means filtering was used for the second denoising. To verify the effectiveness and superiority of the proposed method, experimental verification of simulated data and real data was conducted. Experimental results show that the proposed method can obtain a more ideal SAR interferogram after noise reduction and has better qualitative and quantitative evaluation than the competitive methods, which indicates that the proposed method can achieve effective denoising for SAR interferograms while preserving the detailed features at smaller window settings.

Utility of fast non-local means (FNLM) filter for detection of pulmonary nodules in chest CT for pediatric patient.

This study was aimed to evaluate the utility based on imaging quality of the fast non-local means (FNLM) filter in diagnosing lung nodules in pediatric chest computed tomography (CT). We retrospectively reviewed the chest CT reconstructed with both filtered back projection (FBP) and iterative reconstruction (IR) in pediatric patients with metastatic lung nodules. After applying FNLM filter with six h values (0.0001, 0.001, 0.01, 0.1, 1, and 10) to the FBP images, eight sets of images including FBP, IR, and FNLM were analyzed. The image quality of the lung nodules was evaluated objectively for coefficient of variation (COV), contrast to noise ratio (CNR), and point spread function (PSF), and subjectively for noise, sharpness, artifacts, and diagnostic acceptability. The COV was lowest in IR images and decreased according to increasing h values and highest with FBP images (P<0.001). The CNR was highest with IR images, increased according to increasing h values and lowest with FBP images (P<0.001). The PSF was lower only in FNLM filter with h value of 0.0001 or 0.001 than in IR images (P<0.001). In subjective analysis, only images of FNLM filter with h value of 0.0001 or 0.001 rarely showed unacceptable quality and had comparable results with IR images. There were less artifacts in FNLM images with h value of 0.0001 compared with IR images (p<0.001). FNLM filter with h values of 0.0001 allows comparable image quality with less artifacts compared with IR in diagnosing metastatic lung nodules in pediatric chest CT.

An Adaptive Phase Optimization Algorithm for Distributed Scatterer Phase History Retrieval

The multitemporal interferometric synthetic aperture radar (InSAR) technique based on distributed scatterers (DSs) has been widely applied in high-precision deformation measurements, which compensates for the drawback that the persistent scatterer InSAR technique does not obtain sufficient monitoring points, especially in rural areas. Considering that DS pixels are susceptible to various decorrelation factors, it is necessary to retrieve the optimal phase series by phase optimization algorithms (POAs). However, conventional POAs rely on a sample covariance matrix or complex coherence matrix (CCM) derived by spatially averaging statistically homogeneous pixel neighborhoods, which may blur and destroy phase information, especially in dense fringe areas. To overcome this limitation, an adaptive POA is proposed in this article. The adaptive POA artificially constructs a superior CCM by the filtered interferometric phase, which is derived through spatial adaptive filtering approach fusion of principal phase component estimation and fast nonlocal means filtering, and an accurate coherence matrix determined via coherence estimation bias correction. Moreover, the modified eigen-decomposition-based maximum-likelihood-estimator of the interferometric phase (EMI) with coherence-power-weighting is proposed to further improve the estimation precision and computational efficiency. The estimated CCM is then processed with the modified coherence-power-weighted EMI algorithm, and the optimal phase history is retrieved. The experimental results validated against both simulated and Sentinel-1A data demonstrate the superior optimization performance and robustness of the adaptive POA over traditional POAs.

Feasibility of fast non local means filter in pediatric chest x-ray for increasing of pulmonary nodule detectability with 3D printed lung nodule phantom

General x-ray images have a lower probability of nodule detection than other modalities. Especially in children, the probability of nodule detection can likely drop due to poor image quality from using low radiation dose. To demonstrate the effectiveness of fast non-local means (FNLM) filter to increase the probability of nodule detection in pediatric chest x-ray images and reduce radiation dose while maintaining image quality. Quantitative assessment of normalised noise power spectrum (NNPS), coefficient of variation (COV) and contrast to noise ratio (CNR) were performed after applying four filters (median, Wiener, total variation and FNLM) on a 1-year-old child phantom. A 3D-printed patient nodule phantom was inserted into the phantom. Assessment was performed on AP and LAT view images acquired with the tube voltage reduced to 38 and 27%, and tube current reduced to 84 and 61%, respectively. The results showed the lowest NNPS and COV values and the highest CNR value when the FNLM filter applied. Moreover, the AP view results showed 37% decrease in COV and 30% increase in CNR in images with the FNLM filter applied (images exposed with the tube voltage and current reduced to 29% and 50%, respectively). The LAT view results showed 5% decrease in COV and 36% increase in CNR in images with the FNLM filter applied (images exposed with the tube current reduced by 27%). By applying the FNLM filter, the probability of nodule detection could be increased by denoising and contrast enhancement. Moreover, using the FNLM filter could reduce cancer risk in pediatric patients by reducing radiation dose about 30% to 44%.

A novel two-step noise reduction approach for interferometric phase images

Interferometric phase filtering is a key step in InSAR data processing. However, when utilizing interferometric phase filtering methods, it is difficult to preserve useful phase information while eliminating as many residuals as possible. This paper proposes a novel two-step noise reduction approach for interferometric phase images. This method uses the amended matrix pencil model to extract the main phase component of the interferometric phase in combination with a simple and effective adaptive window determination method for the first time. Then, fast non-local means filtering is used to extract an accurate residual phase component. Finally, the principal phase component extracted in the first step is summed with the residual component from the second step to obtain the final result. Several common interferometric phase filtering methods are compared using two sets of simulated data and one set of real TanDEM-X InSAR data. The results show that the proposed method can not only preserve useful phase information but also effectively remove residuals. In addition, the proposed method boasts a superior filtering efficiency.

Quantitative study of fast non-local means-based denoising filter in chest X-ray imaging with lung nodule using three-dimensional printing

In chest radiography, a solitary pulmonary nodule, which may be a precursor of lung cancer, is a frequently detected finding. However, as the image quality is deteriorated owing to the increase in the noise, lung cancer screening studies revealed that the likelihood of finding a nodule is lower than those of other modalities. This study quantitatively evaluates three widely used filters (median, Wiener, and total variation) and a newly proposed filter (fast non-local means (FNLM)), which reduce image noise. Images of a phantom with lung nodules, obtained from a patient using the 3D printing technology, were acquired at the chest anterior–posterior, lateral, and posterior–anterior positions. To evaluate their denoising performance, normalized noise power spectrum, contrast to noise ratio and coefficient of variation were used. In the quantitative evaluation of the overall image, the proposed FNLM filter exhibited the best image performance. In the quantitative evaluation of the nodule image, the FNLM filter, which exhibits outstanding denoising performance and time efficiency, can be employed. Therefore, with the use of the FNLM filter in chest radiography, the detection probability of a nodule, which can be a precursor of lung cancer, is increased, and the cancer can be prevented even with a lower dose.

Two dimensional cuckoo search optimization algorithm based despeckling filter for the real ultrasound images

A clinical ultrasound imaging plays a significant role in the proper diagnosis of patients because, it is a cost-effective and non-invasive technique in comparison with other methods. The speckle noise contamination caused by ultrasound images during the acquisition process degrades its visual quality, which makes the diagnosis task difficult for physicians. Hence, to improve their visual quality, despeckling filters are commonly used for processing of such images. However, several disadvantages of existing despeckling filters discourage the use of existing despeckling filters to reduce the effect of speckle noise. In this paper, two dimensional cuckoo search optimization algorithm based despeckling filter is proposed for avoiding limitations of various existing despeckling filters. Proposed despeckling filter is developed by combining fast non-local means filter and 2D finite impulse response (FIR) filter with cuckoo search optimization algorithm. In the proposed despeckling filter, the coefficients of 2D FIR filter are optimized by using the cuckoo search optimization algorithm. The quantitative results comparison between the proposed despeckling filter and other existing despeckling filters are analyzed by evaluating PSNR, MSE, MAE, and SSIM values for different real ultrasound images. Results reveal that the visual quality obtained by the proposed despeckling filter is better than other existing despeckling filters. The numerical results also reveal that the proposed despeckling filter is highly effective for despeckling the clinical ultrasound images.

Feasibility of newly designed fast non local means (FNLM)-based noise reduction filter for X-ray imaging: A simulation study

In the diagnostic radiology field, reducing the radiation dose for patient lead to increased noise in image. Since increases of noise decrease the diagnosis rate, to reduce the noise is necessary. In this study quantitatively evaluates the four widely used and newly verified filters which remove noise in image: median, Wiener, total variation, and fast non local means (FNLM). For that purpose, X-ray and computed tomography (CT) images are acquired using MATLAB simulation with 3D voxelized phantom. To evaluate image performance, normalized noise power spectrum (NNPS), contrast to noise ratio (CNR) and coefficient of variation (COV) were used. As a result, we can efficiently remove noise in X-ray image when FNLM filter was used compared with frequently used filters. In conclusion, our results demonstrated that our proposed FNLM filter shows superior denoising performance, which is expected to enhance the detection of diseases in clinical images with low dose.

GPU-Based Block-Wise Nonlocal Means Denoising for 3D Ultrasound Images

Speckle suppression plays an important role in improving ultrasound (US) image quality. While lots of algorithms have been proposed for 2D US image denoising with remarkable filtering quality, there is relatively less work done on 3D ultrasound speckle suppression, where the whole volume data rather than just one frame needs to be considered. Then, the most crucial problem with 3D US denoising is that the computational complexity increases tremendously. The nonlocal means (NLM) provides an effective method for speckle suppression in US images. In this paper, a programmable graphic-processor-unit- (GPU-) based fast NLM filter is proposed for 3D ultrasound speckle reduction. A Gamma distribution noise model, which is able to reliably capture image statistics for Log-compressed ultrasound images, was used for the 3D block-wise NLM filter on basis of Bayesian framework. The most significant aspect of our method was the adopting of powerful data-parallel computing capability of GPU to improve the overall efficiency. Experimental results demonstrate that the proposed method can enormously accelerate the algorithm.