Search

The Non-Local Means (NLM) algorithm is a fundamental denoising technique widely utilized in various domains of image processing. However, further research is essential to gain a comprehensive understanding of its capabilities and limitations. This includes determining the types of noise it can effectively remove, choosing an appropriate kernel, and assessing its convergence behavior. In this study, we optimize the NLM algorithm for all variations of independent and identically distributed (i.i.d.) variance-stabilized noise and conduct a thorough examination of its convergence behavior. We introduce the concept of the optimal oracle NLM, which minimizes the upper bound of pointwise L 1 or L 2 risk. We demonstrate that the optimal oracle weights comprise triangular kernels with point-adaptive bandwidth, contrasting with the commonly used Gaussian kernel, which has a fixed bandwidth. The computable optimal weighted NLM is derived from this oracle filter by replacing the similarity function with an estimator based on the similarity patch. We present theorems demonstrating that both the oracle filter and the computable filter achieve optimal convergence rates under minimal regularity conditions. Finally, we conduct numerical experiments to validate the performance, accuracy, and convergence of L 1 and L 2 risk minimization for NLM. These convergence theorems provide a theoretical foundation for further advancing the study of the NLM algorithm and its practical applications. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Test & Measurement Technology is the property of Publishing Center of North University of China and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Real-time monitoring of heartbeat signals using multimode fiber optic microvibration sensing technology is crucial for diagnosing cardiovascular diseases, but the heartbeat signals are very weak and susceptible to noise interference, leading to inaccurate diagnostic results. In this paper, a combined enhanced discrete wavelet transform (DWT) and non-local mean estimation (NLM) denoising method is proposed to remove noise from heartbeat signals, which adaptively determines the filtering parameters of the DWT-NLM composite method using objective noise reduction quality assessment metrics by denoising different ECG signals from multiple databases with the addition of additive Gaussian white noise (AGW) with different signal-to-noise ratios (SNRs). The noise reduction results are compared with those of NLM, enhanced DWT, and conventional DWT combined with NLM method. The results show that the output SNR of the proposed method is significantly higher than the other methods compared in the range of −5 to 25 dB input SNR. Further, the proposed method is employed for noise reduction of heartbeat signals measured by fiber optic microvibration sensing. It is worth mentioning that the proposed method does not need to obtain the exact noise level, but only the adaptive filtering parameters based on the autocorrelation nature of the denoised signal. This work greatly improves the signal quality of the multimode fiber microvibration sensing system and helps to improve the diagnostic accuracy. [ABSTRACT FROM AUTHOR]

Medical image denoising plays a critical role in enhancing the quality of diagnostic imaging, where noise reduction without compromising image details is paramount. In this paper, we propose a novel hybrid approach aimed at improving the denoising efficacy for medical images. Initially, we employ an adaptive Kalman filter to attenuate noise, leveraging its proficiency in state estimation from noisy measurements. Unlike conventional Kalman filters with fixed parameters, our adaptive Kalman filter dynamically adjusts its parameters based on the noise characteristics of the input image, thus offering enhanced accuracy in estimating the underlying true state of the system represented by the medical image. Subsequently, both a non-local means (NLM) method and a median filter are introduced as post-processing steps to further refine the denoised image. The NLM method leverages the similarities between image patches to effectively reduce noise, while the median filter further enhances the denoised image by suppressing residual noise and preserving image details. However, the effectiveness of NLM and the median filter is highly dependent on carefully chosen parameters, which traditionally necessitates extensive computational resources for optimization. To address this challenge, we introduce the innovative use of Latin square optimization, a structured experimental design technique, to efficiently determine optimal parameters for NLM. By systematically exploring parameter combinations using Latin square optimization, we mitigate the complexity of experiments while enhancing denoising performance. The experimental results on medical images demonstrate the effectiveness of our proposed approach, showcasing significant improvements in noise reduction and the preservation of image features compared to conventional methods. Our hybrid approach not only advances the state-of-the-art in medical image denoising but also presents a practical solution for optimizing parameter selection in NLM, thereby facilitating their broader adoption in medical imaging applications. [ABSTRACT FROM AUTHOR]

Advances in medical imaging modalities facilitate the early and accurate detection of tumors of various types. A preferred imaging modality for diagnosis and identification of tumors is the B-mode ultrasound imaging, but due to the noise and artifacts present, correct interpretation of lesions region becomes a difficult task for an inexperienced radiologist. In this context, an efficient and reliable computer-aided segmentation system is preferred for extracting regions of interest. Recently, conventional methods of segmentation have been replaced by deep learning methods. In this article, a novel Multi-Residual U-Net model is proposed for the segmentation of ultrasound medical images. This architecture adopts residual blocks to improve the performance of deep convolutional networks and a loss function that addresses the class imbalance issue. To improve the quality and reduce Speckle noise, input images are pre-processed using an optimized Non-Local Means filter. Three benchmark B-mode Ultrasound image datasets of 200 Breast lesion images, 504 Skeletal images, and 647 Breast Lesion images are used for experimentation. Experimental results demonstrate that the proposed model performs more accurate segmentation in comparison to the five deep models chosen for the study. [ABSTRACT FROM AUTHOR]

Image denoise has been explored with the development of various filters used to remove or reduce random disruptions on observed data, but at the same time it preserves most of the edges and the fine details of the scene. The issue caused by the combined deterioration of the Gaussian noise succeeds the scattering through all the signal frequencies. Thus, the most effective filters for this type of noise are implemented in spatial domain. In this article, we proposed a Non-Local Means filter that combines the average of each fragment of a browser window, by using four measures – of distinct similarities – among the Gaussian densities that are estimated from the following fragments: the Kullback-Leibler divergence, the Bhattacharyya distance, the Hellinger distance and the Cauchy-Schwarz divergence. Computational experiments were done in a set of 7 images that were deteriorated by a noise of Gaussian type, considering that the data obtained show that the proposed methods are capable of producing, on average, a Peak Signal-to-Noise Ratio significantly greater than the one the combination of Total Variation, Non-Local Means, BM3D, Anisotropic Diffusion, Wiener, Wavelet e Bilateral filters does when they are applied independently. [ABSTRACT FROM AUTHOR]

The three-dimensional (3D) geometrical information that depth maps contain is useful in many applications such as 3D reconstruction or simultaneous localization and mapping (SLAM). Kinect is widely used in depth image acquisition due to its low cost and good real-time performance. However, the quality of depth images obtained by Kinect is influenced by holes which make depth image inadequate for further applications. To suppress the influence of holes on a subsequent application, a fixing algorithm of Kinect depth image based on non-local means (NLM) is proposed in this paper. The holes in depth image are filled using the weights which are calculated on the corresponding gray image by distance factor and value consistent factor. And the experiment results demonstrate that the proposed method achieves good performance in both evaluation in metrics and subjectively visual effect. This research provides a solution idea for depth image fixing algorithm with low complexity. [ABSTRACT FROM AUTHOR]

RGB-D cameras provide depth and color information and are widely used in 3D reconstruction and computer vision. In the majority of existing RGB-D cameras, a considerable portion of depth values is often lost due to severe occlusion or limited camera coverage, thereby adversely impacting the precise localization and three-dimensional reconstruction of objects. In this paper, to address the issue of poor-quality in-depth images captured by RGB-D cameras, a depth image hole repair algorithm based on non-local means is proposed first, leveraging the structural similarities between grayscale and depth images. Second, while considering the cumbersome parameter tuning associated with the non-local means hole repair method for determining the size of structural blocks for depth image hole repair, an intelligent block factor is introduced, which automatically determines the optimal search and repair block sizes for various hole sizes, resulting in the development of an adaptive block-based non-local means algorithm for repairing depth image holes. Furthermore, the proposed algorithm's performance are evaluated using both the Middlebury stereo matching dataset and a self-constructed RGB-D dataset, with performance assessment being carried out by comparing the algorithm against other methods using five metrics: RMSE, SSIM, PSNR, DE, and ALME. Finally, experimental results unequivocally demonstrate the innovative resolution of the parameter tuning complexity inherent in-depth image hole repair, effectively filling the holes, suppressing noise within depth images, enhancing image quality, and achieving elevated precision and accuracy, as affirmed by the attained results. [ABSTRACT FROM AUTHOR]

BackgroundElectrocardiogram (ECG) signals are inevitably contaminated with various kinds of noises during acquisition and transmission. The presence of noises may produce the inappropriate information on cardiac health, thereby preventing specialists from making correct analysis.MethodsIn this paper, an efficient strategy is proposed to denoise ECG signals, which employs a time-frequency framework based on S-transform (ST) and combines bi-dimensional empirical mode decomposition (BEMD) and non-local means (NLM). In the method, the ST maps an ECG signal into a subspace in the time frequency domain, then the BEMD decomposes the ST-based time-frequency representation (TFR) into a series of sub-TFRs at different scales, finally the NLM removes noise and restores ECG signal characteristics based on structural self-similarity.ResultsThe proposed method is validated using numerous ECG signals from the MIT-BIH arrhythmia database, and several different types of noises with varying signal-to-noise (SNR) are taken into account. The experimental results show that the proposed technique is superior to the existing wavelet based approach and NLM filtering, with the higher SNR and structure similarity index measure (SSIM), the lower root mean squared error (RMSE) and percent root mean square difference (PRD).ConclusionsThe proposed method not only significantly suppresses the noise presented in ECG signals, but also preserves the characteristics of ECG signals better, thus, it is more suitable for ECG signals processing.

Real-time monitoring of heartbeat signals using multimode fiber optic microvibration sensing technology is crucial for diagnosing cardiovascular diseases, but the heartbeat signals are very weak and susceptible to noise interference, leading to inaccurate diagnostic results. In this paper, a combined enhanced discrete wavelet transform (DWT) and non-local mean estimation (NLM) denoising method is proposed to remove noise from heartbeat signals, which adaptively determines the filtering parameters of the DWT-NLM composite method using objective noise reduction quality assessment metrics by denoising different ECG signals from multiple databases with the addition of additive Gaussian white noise (AGW) with different signal-to-noise ratios (SNRs). The noise reduction results are compared with those of NLM, enhanced DWT, and conventional DWT combined with NLM method. The results show that the output SNR of the proposed method is significantly higher than the other methods compared in the range of −5 to 25 dB input SNR. Further, the proposed method is employed for noise reduction of heartbeat signals measured by fiber optic microvibration sensing. It is worth mentioning that the proposed method does not need to obtain the exact noise level, but only the adaptive filtering parameters based on the autocorrelation nature of the denoised signal. This work greatly improves the signal quality of the multimode fiber microvibration sensing system and helps to improve the diagnostic accuracy.

The accuracy of pipeline temperature monitoring using the Brillouin Optical Time Domain Analysis system depends on the Brillouin Gain Spectrum in the Brillouin Optical Time Domain Analysis system. The Non-Local Means noise reduction algorithm, due to its ability to use the data patterns available within the two-dimensional measurement data space, has been used to improve the Brillouin Gain Spectrum in the Brillouin Optical Time Domain Analysis system. This paper studies a new Non-Local Means algorithm optimized through the Black Widow Optimization Algorithm, in view of the unreasonable selection of smoothing parameters in other Non-Local Means algorithms. The field test demonstrates that, the new algorithm, when compared to other Non-Local Means methods, excels in preserving the detailed information within the Brillouin Gain Spectrum. It successfully restores the fundamental shape and essential characteristics of the Brillouin Gain Spectrum. Notably, at the 25 km fiber end, it achieves a 3 dB higher Signal-to-Noise Ratio compared to other Non-Local Means noise reduction algorithms. Furthermore, the Brillouin Gain Spectrum values exhibit increases of 9.4% in Root Mean Square Error, 12.5% in Sum of Squares Error, and 10% in Full Width at Half Maximum. The improved method has a better denoising effect and broad application prospects in pipeline safety. [ABSTRACT FROM AUTHOR]

Preserving edges while denoising an image is a crucial and challenging necessity. In this paper, a noise-aided edge-preserving denoising algorithm is proposed by extending the classical patch-based Non Local Means (NLM) algorithm. The classical NLM algorithm uses similarity weights to ensure edge preservation. In the proposed algorithm, these similarity weights are enhanced by processing the similarity distances using stochastic resonance (SR). SR refers to a phenomenon where the performance metric in a nonlinear system counterintuitively increases to a peak and then decreases (like a bell curve) in the presence of a controlled amount of noise. The similarity weights derived from NLM are iteratively processed using the discretized SR equation. For the iteratively reconstructed images, the local maxima of the corresponding quality metric, PSNR, is selected as the optimal output. The iterative processing results in a nonlinear scaling of the similarity distances. This processing, in effect, ensures that similarity weights of similar patches are high and those of dissimilar patches are low, thereby producing enhanced edge preservation. The performance of the proposed algorithm is demonstrated by presenting the comparative results for a variety of images corrupted by a wide range of AWGN noise. The proposed algorithm is found to handle the spurious artifacts near the edges more efficiently. Moreover, the obtained edges are sharper and better preserved even in the presence of high noise deviation. Benchmarking results on SET12 and BSD68 datasets show an improvement of 14.5% and 12.1% respectively over that of NLM for high noise. [ABSTRACT FROM AUTHOR]

Copyright of China Rural Water & Hydropower is the property of China Rural Water & Hydropower Editorial Office and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Low-dose computed tomography (LDCT) can effectively reduce radiation exposure in patients. However, with such dose reductions, large increases in speckled noise and streak artifacts occur, resulting in seriously degraded reconstructed images. The non-local means (NLM) method has shown potential for improving the quality of LDCT images. In the NLM method, similar blocks are obtained using fixed directions over a fixed range. However, the denoising performance of this method is limited. In this paper, a region-adaptive NLM method is proposed for LDCT image denoising. In the proposed method, pixels are classified into different regions according to the edge information of the image. Based on the classification results, the adaptive searching window, block size and filter smoothing parameter could be modified in different regions. Furthermore, the candidate pixels in the searching window could be filtered based on the classification results. In addition, the filter parameter could be adjusted adaptively based on intuitionistic fuzzy divergence (IFD). The experimental results showed that the proposed method performed better in LDCT image denoising than several of the related denoising methods in terms of numerical results and visual quality.

Diffusion tensor imaging (DTI) is an advanced magnetic resonance technology that describes subtle brain structures using a diffusion tensor at each point. The obtained DTI image is always degraded since diffusion-weighted imaging sequences, which are used to estimate DTI images, are corrupted by noise. In this paper, we propose an approach called geometric invariant nonlocal means on the tensor manifold (GINLM-TM) to reduce undesired components in the degraded DTI image. We transform the diffusion tensor into a positive definite matrix (called a tensor) to measure the intrinsic property of the diffusion tensor. Then, we directly regularise DTI images in the tensor manifold endowed with an affine invariant metric. Finally, geometrically invariant measures of patches of tensors are used to define the similarity function of patches to ensure the similarity between patches is more accurate and robust. It is experimentally demonstrated that the proposed method performs adequately in reducing undesired components without blurring the boundaries of DTI images. The results of fractional anisotropy (FA) images and fibre tracking of our restored data indicate that our method performs well in denoising the DTI image. [ABSTRACT FROM AUTHOR]

Non-local means (NLM) remove noise from an image by a weighted average process. Its distinctive feature is that the weight of a pixel depends on the similarity between two image patches. Patch similarity is measured by the Euclidean distance between two patches and pixels with smaller mutual distance have larger weights. Because of these contextually related weights, image structure can be much better preserved for NLM. Therefore, patch distance is paramount to the denoising performance of NLM. In this paper, we show that a correlation term previously omitted in theoretical analysis cannot be ignored in patch distance calculation, especially when noise is large. We then propose to add a correlation correction function in the distance calculation to remedy this omission. Experimental results show that both quantitative and qualitative performance of the correlation corrected NLM is significantly improved as compared to the other NLM algorithms under large noise. The correlation corrected NLM can also be competitive with the state-of-the-art block-matching 3D (BM3D) algorithm and its execution time is much shorter than BM3D. [ABSTRACT FROM AUTHOR]

RGB-D cameras provide depth and color information and are widely used in 3D reconstruction and computer vision. In the majority of existing RGB-D cameras, a considerable portion of depth values is often lost due to severe occlusion or limited camera coverage, thereby adversely impacting the precise localization and three-dimensional reconstruction of objects. In this paper, to address the issue of poor-quality in-depth images captured by RGB-D cameras, a depth image hole repair algorithm based on non-local means is proposed first, leveraging the structural similarities between grayscale and depth images. Second, while considering the cumbersome parameter tuning associated with the non-local means hole repair method for determining the size of structural blocks for depth image hole repair, an intelligent block factor is introduced, which automatically determines the optimal search and repair block sizes for various hole sizes, resulting in the development of an adaptive block-based non-local means algorithm for repairing depth image holes. Furthermore, the proposed algorithm’s performance are evaluated using both the Middlebury stereo matching dataset and a self-constructed RGB-D dataset, with performance assessment being carried out by comparing the algorithm against other methods using five metrics: RMSE, SSIM, PSNR, DE, and ALME. Finally, experimental results unequivocally demonstrate the innovative resolution of the parameter tuning complexity inherent in-depth image hole repair, effectively filling the holes, suppressing noise within depth images, enhancing image quality, and achieving elevated precision and accuracy, as affirmed by the attained results.

The accuracy of pipeline temperature monitoring using the Brillouin Optical Time Domain Analysis system depends on the Brillouin Gain Spectrum in the Brillouin Optical Time Domain Analysis system. The Non-Local Means noise reduction algorithm, due to its ability to use the data patterns available within the two-dimensional measurement data space, has been used to improve the Brillouin Gain Spectrum in the Brillouin Optical Time Domain Analysis system. This paper studies a new Non-Local Means algorithm optimized through the Black Widow Optimization Algorithm, in view of the unreasonable selection of smoothing parameters in other Non-Local Means algorithms. The field test demonstrates that, the new algorithm, when compared to other Non-Local Means methods, excels in preserving the detailed information within the Brillouin Gain Spectrum. It successfully restores the fundamental shape and essential characteristics of the Brillouin Gain Spectrum. Notably, at the 25 km fiber end, it achieves a 3 dB higher Signal-to-Noise Ratio compared to other Non-Local Means noise reduction algorithms. Furthermore, the Brillouin Gain Spectrum values exhibit increases of 9.4% in Root Mean Square Error, 12.5% in Sum of Squares Error, and 10% in Full Width at Half Maximum. The improved method has a better denoising effect and broad application prospects in pipeline safety.

High-dynamic-range (HDR) image reconstruction methods are designed to fuse multiple Low-dynamic-range (LDR) images captured with different exposure values into a single HDR image. Recent CNN-based methods mostly perform local attention- or alignment-based fusion of multiple LDR images to create HDR contents. Depending on a single attention mechanism or alignment causes failure in compensating ghosting artifacts, which can arise in the synthesized HDR images due to the motion of objects or camera movement across different LDR image inputs. In this study, we propose a multi-scale attention-guided non-local network called MSANLnet for efficient HDR image reconstruction. To mitigate the ghosting artifacts, the proposed MSANLnet performs implicit alignment of LDR image features with multi-scale spatial attention modules and then reconstructs pixel intensity values using long-range dependencies through non-local means-based fusion. These modules adaptively select useful information that is not damaged by an object's movement or unfavorable lighting conditions for image pixel fusion. Quantitative evaluations against several current state-of-the-art methods show that the proposed approach achieves higher performance than the existing methods. Moreover, comparative visual results show the effectiveness of the proposed method in restoring saturated information from original input images and mitigating ghosting artifacts caused by large movement of objects. Ablation studies show the effectiveness of the proposed method, architectural choices, and modules for efficient HDR reconstruction. [ABSTRACT FROM AUTHOR]

Parameter setting and information redundancy are essential issues for the non-local means (NLM) algorithm. This paper introduces a new factor based on the Hessian matrix to adapt the smoothing parameter. Then, a strategy is proposed to implement the NLM by representing patches in terms of features, which uses the 2D histogram and summed-area table. Compared with other methods, the metric for patch similarity in this paper is based on statistical features of patches instead of Euclidean distance. More importantly, not many predefined thresholds are needed. Experimental results show that the proposed algorithm obtains better visual quality and numerical results, especially for images with rich contents and high noise. [ABSTRACT FROM AUTHOR]

Osteosarcoma is the most common malignant osteosarcoma, and most developing countries face great challenges in the diagnosis due to the lack of medical resources. Magnetic resonance imaging (MRI) has always been an important tool for the detection of osteosarcoma, but it is a time-consuming and labor-intensive task for doctors to manually identify MRI images. It is highly subjective and prone to misdiagnosis. Existing computer-aided diagnosis methods of osteosarcoma MRI images focus only on accuracy, ignoring the lack of computing resources in developing countries. In addition, the large amount of redundant and noisy data generated during imaging should also be considered. To alleviate the inefficiency of osteosarcoma diagnosis faced by developing countries, this paper proposed an artificial intelligence multiprocessing scheme for pre-screening, noise reduction, and segmentation of osteosarcoma MRI images. For pre-screening, we propose the Slide Block Filter to remove useless images. Next, we introduced a fast non-local means algorithm using integral images to denoise noisy images. We then segmented the filtered and denoised MRI images using a U-shaped network (ETUNet) embedded with a transformer layer, which enhances the functionality and robustness of the traditional U-shaped architecture. Finally, we further optimized the segmented tumor boundaries using conditional random fields. This paper conducted experiments on more than 70,000 MRI images of osteosarcoma from three hospitals in China. The experimental results show that our proposed methods have good results and better performance in pre-screening, noise reduction, and segmentation. [ABSTRACT FROM AUTHOR]

Purpose. Cherenkov imaging during radiotherapy provides a real time visualization of beam delivery on patient tissue, which can be used dynamically for incident detection or to review a summary of the delivered surface signal for treatment verification. Very few photons form the images, and one limitation is that the noise level per frame can be quite high, and mottle in the cumulative processed images can cause mild overall noise. This work focused on removing or suppressing noise via image postprocessing. Approach. Images were analyzed for peak-signal-to-noise and spatial frequencies present, and several established noise/mottle reduction algorithms were chosen based upon these observations. These included total variation minimization (TV-L1), non-local means filter (NLM), block-matching 3D (BM3D), alpha (adaptive) trimmed mean (ATM), and bilateral filtering. Each were applied to images acquired using a BeamSite camera (DoseOptics) imaged signal from 6x photons from a TrueBeam linac delivering dose at 600 MU min -1 incident on an anthropomorphic phantom and tissue slab phantom in various configurations and beam angles. The standard denoised images were tested for PSNR, noise power spectrum (NPS) and image sharpness. Results. The average peak-signal-to-noise ratio (PSNR) increase was 17.4% for TV-L1. NLM denoising increased the average PSNR by 19.1%, BM3D processing increased it by12.1% and the bilateral filter increased the average PSNR by 19.0%. Lastly, the ATM filter resulted in the lowest average PSNR increase of 10.9%. Of all of these, the NLM and bilateral filters produced improved edge sharpness with, generally, the lowest NPS curve. Conclusion. For cumulative image Cherenkov data, NLM and the bilateral filter yielded optimal denoising with the TV-L1 algorithm giving comparable results. Single video frame Cherenkov images exhibit much higher noise levels compared to cumulative images. Noise suppression algorithms for these frame rates will likely be a different processing pipeline involving these filters incorporated with machine learning., (Creative Commons Attribution license.)

Abstract Non-local Means (NLMs) play essential roles in image denoising, restoration, inpainting, etc., due to its simple theory but effective performance. However, when the noise increases, the denoising accuracy of NLMs decreases significantly. This paper further develop the NLMs-based denoising method to remove noise with less loss of image details. It is realized by embedding an optimal graph edge weights driven NLMs kernel into a multi-layer residual compensation framework. Unlike the patch similarity-based weights in the traditional NLMs filters, the edge weights derived from the optimal graph Laplacian regularization consider (1) the distance between the target pixel and the candidate pixel, (2) the local gradient and (3) the patch similarity. After defining the weights, the graph-based NLMs kernel is then put into a multi-layer framework. The corresponding primal and residual terms at each layer are finally fused with learned weights to recover the image. Experimental results show that our method is effective and robust, especially for piecewise smooth images.

Human pose estimation is the basic task in computer vision. It can be used in action recognition, games, animation production and so on. Inspired by the nonlocal mean method, a non-local high-resolution (NLHR) network is designed. Fusion of non-local network modules in the network phase of the original image 1/32 resolution, so that the network has the ability to obtain global features, in order to improve the accuracy of human posture estimation. The NLHR network is trained on MPII datasets and tested on the MPII validation set, the average accuracy under PCKh@0.5 evaluation criteria is 90.5%, 0.2 percentage points higher than the HRNet baseline. Training on the COCO human key point detection dataset, test result on the COCO validation set AP is 76.7%, 2.3 percentage points higher than HRNet baseline. Three ablation experiments verify that the NLHR network for human pose estimation can exceed the existing human pose estimation network in accuracy. [ABSTRACT FROM AUTHOR]

The existence of multiplicative noise in synthetic aperture radar (SAR) images makes SAR segmentation by fuzzy c-means (FCM) a challenging task. To cope with speckle noise, we first propose an unsupervised FCM with embedding log-transformed Bayesian non-local spatial information (LBNL_FCM). This non-local information is measured by a modified Bayesian similarity metric which is derived by applying the log-transformed SAR distribution to Bayesian theory. After, we construct the similarity metric of patches as the continued product of corresponding pixel similarity measured by generalized likelihood ratio (GLR) to avoid the undesirable characteristics of log-transformed Bayesian similarity metric. An alternative unsupervised FCM framework named GLR_FCM is then proposed. In both frameworks, an adaptive factor based on the local intensity entropy is employed to balance the original and non-local spatial information. Additionally, the membership degree smoothing and the majority voting idea are integrated as supplementary local information to optimize segmentation. Concerning experiments on simulated SAR images, both frameworks can achieve segmentation accuracy of over 97%. On real SAR images, both unsupervised FCM segmentation frameworks work well on SAR homogeneous segmentation in terms of region consistency and edge preservation. [ABSTRACT FROM AUTHOR]

The ultrasonic technique is an indispensable imaging modality for diagnosis of breast cancer in young women due to its ability in efficiently capturing the tissue properties, and decreasing nega-tive recognition rate thereby avoiding non-essential biopsies. Despite the advantages, ultrasound images are affected by speckle noise, generating fine-false structures that decrease the contrast of the images and diminish the actual boundaries of tissues on ultrasound image. Moreover, speckle noise negatively impacts the subsequent stages in image processing pipeline, such as edge detec-tion, segmentation, feature extraction, and classification. Previous studies have formulated vari-ous speckle reduction methods in ultrasound images; however, these methods suffer from being unable to retain finer edge details and require more processing time. In this study, we propose a breast ultrasound de-speckling method based on rotational invariant block matching non-local means (RIBM-NLM) filtering. The effectiveness of our method has been demonstrated by com-paring our results with three established de-speckling techniques, the switching bilateral filter (SBF), the non-local means filter (NLMF), and the optimized non-local means filter (ONLMF) on 250 images from public dataset and 6 images from private dataset. Evaluation metrics, including Self-Similarity Index Measure (SSIM), Peak Signal to Noise Ratio (PSNR), and Mean Square Error (MSE) were utilized to measure performance. With the proposed method, we were able to record average SSIM of 0.8915, PSNR of 65.97, MSE of 0.014, RMSE of 0.119, and computational speed of 82 seconds at noise variance of 20dB using the public dataset, all with p-value of less than 0.001 compared against NLMF, ONLMF, and SBF. Similarly, the proposed method achieved av-erage SSIM of 0.83, PSNR of 66.26, MSE of 0.015, RMSE of 0.124, and computational speed of 83 seconds at noise variance of 20dB using the private dataset, all with p-value of less than 0.001 compared against NLMF, ONLMF, and SBF. [ABSTRACT FROM AUTHOR]

Many methods have been proposed for image denoising, among which the non-local means (NLM) denoising is widely used for fully exploiting the self-similarity of natural images. For NLM denoising, it needs to calculate the similarity of clean image blocks as weights. But affected by noise, it is challenging to accurately get the similarity of clean image patches. Most existing NLM denoising approaches often cause the restored image to be over smoothed and lose lots of details, especially for image with high noise levels. To tackle this, a novel singular value decomposition-based similarity measure method is proposed, which can effectively reduce the disturbance of noise. For the method, we first calculate and vectorize the singular values of two image patches extracted from the noisy image and compute Euclidean distances and cross-angles of the vectors. Second, we propose to utilize the geometric average of Euclidean distance and cross-angle to calculate the similarity between two image patches which is resistant to noise. Third, the proposed similarity measure is applied to non-local means denoising to compute similarity of noisy image patches. Experimental results show that compared to state-of-the-art denoising algorithms, the proposed method can effectively eliminate noise and restore more details with higher peak signal-to-noise ratio (PSNR) and the structural similarity (SSIM) index values. [ABSTRACT FROM AUTHOR]

While self-attention has been successfully applied in a variety of natural language processing and computer vision tasks, its application in Monte Carlo (MC) image denoising has not yet been well explored. This paper presents a self-attention based MC denoising deep learning network based on the fact that self-attention is essentially non-local means filtering in the embedding space which makes it inherently very suitable for the denoising task. Particularly, we modify the standard self-attention mechanism to an auxiliary feature guided self-attention that considers the by-products (e.g., auxiliary feature buffers) of the MC rendering process. As a critical prerequisite to fully exploit the performance of self-attention, we design a multi-scale feature extraction stage, which provides a rich set of raw features for the later self-attention module. As self-attention poses a high computational complexity, we describe several ways that accelerate it. Ablation experiments validate the necessity and effectiveness of the above design choices. Comparison experiments show that the proposed self-attention based MC denoising method outperforms the current state-of-the-art methods. [ABSTRACT FROM AUTHOR]

Image reconstruction from noisy data has a long history of methodological development and is based on a variety of ideas. In this paper we introduce a new method called patchwise adaptive smoothing, that extends the propagation-separation approach by using comparisons of local patches of image intensities to define local adaptive weighting schemes for an improved balance of reduced variability and bias in the reconstruction result. We present the implementation of the new method in an R package aws and demonstrate its properties on a number of examples in comparison with other state-of-the art image reconstruction methods.

Electrocardiogram (ECG) is a critical biological signal, which usually carries a great deal of essential information about patients. The high quality ECG signals are always required for a proper diagnosis of cardiac disorders. However, the raw ECG signals are highly noisy in nature. In the paper, we propose a hybrid denoising scheme to enhance ECG signals by combining high-order synchrosqueezing transform (FSSTH) with non-local means (NLM). With this method, a noisy ECG signal is first decomposed into an ensemble of intrinsic mode functions (IMFs) by FSSTH. Then, some noise is removed by eliminating a set of noisy IMFs that are determined by a scaling exponent obtained by the detrended fluctuation analysis (DFA); while the remaining IMFs are filtered by NLM. Finally, the denoised ECG signal is obtained by reconstructing the processed IMFs. Experiments are carried out using the simulated ECG signals and real ones from the MIT-BIH database, and the denoising performances are evaluated in terms of signal to noise ratio (SNR), root mean squared error (RMSE) and percent root mean square difference (PRD). Results show that the hybrid denoising scheme involving both FSSTH and NLM is able to suppress complex noise from ECG signals more effectively while preserving the details well.

The topic of denoising magnetic resonance (MR) images is considered in this paper. More in detail, an enhanced Non-Local Means (NLM) filter using the Kolmogorov-Smirnov (KS) distance is proposed. The KS-NLM approach estimates the similarity between image patches by computing the KS distance. To overcome that NLM filters assign the same role to all pixels in patches, that is, not privileging the central one, we propose a new filter, namely the Anisotropic Weighted KS-NLM (Aw KS-NLM), which better deals with central pixels within the patches by, on one hand, including a suitable weighted strategy and, on the other, by performing a local anisotropy analysis. The Aw KS-NLM has been compared to other existing non-local Means (NLM) methodologies in both MRI simulated and real datasets. The results provide excellent noise reduction and image-detail preservation.

Urban vegetation inventory at city-scale using terrestrial light detection and ranging (LiDAR) point clouds is very challenging due to the large quantity of points, varying local density, and occlusion effects, leading to missing features and incompleteness of data. This paper proposes a novel method, named Point Non-Local Means (PointNLM) network, which incorporates the supervoxel-based and point-wise for automatic semantic segmentation of vegetation from large scale complex scene point clouds. PointNLM captures the long-range relationship between groups of points via a non-local branch cascaded three times to describe sharp geometric features. Simultaneously, a local branch processes the position of scattered feature points and captures the low and high level features. Finally, we proposed a fusion layer of neighborhood max-pooling method to concatenate the long-range features, low level features and high level features for segmenting the trees. The proposed architecture was evaluated on three datasets, including two open access datasets of Semantic3D and Paris-Lille-3D, and an in-house dataset acquired by a commercial mobile LiDAR system. Experimental results indicated that the proposed method provides an efficient and robust result for vegetation segmentation, achieving an Intersection over Union (IoU) of 94.4%, F1-score of 92.7% and overall accuracy of 96.3%, respectively.

The GF-3 satellite is China’s first self-developed active imaging C-band multi-polarization synthetic aperture radar (SAR) satellite with complete intellectual property rights, which is widely used in various fields. Among them, the detection and recognition of banklines of GF-3 SAR image has very important application value for map matching, ship navigation, water environment monitoring and other fields. However, due to the coherent imaging mechanism, the GF-3 SAR image has obvious speckle, which affects the interpretation of the image seriously. Based on the excellent multi-scale, directionality and the optimal sparsity of the shearlet, a bankline detection algorithm based on shearlet is proposed. Firstly, we use non-local means filter to preprocess GF-3 SAR image, so as to reduce the interference of speckle on bankline detection. Secondly, shearlet is used to detect the bankline of the image. Finally, morphological processing is used to refine the bankline and further eliminate the false bankline caused by the speckle, so as to obtain the ideal bankline detection results. Experimental results show that the proposed method can effectively overcome the interference of speckle, and can detect the bankline information of GF-3 SAR image completely and smoothly.

Non-local Means (NLMs) play essential roles in image denoising, restoration, inpainting, etc., due to its simple theory but effective performance. However, when the noise increases, the denoising accuracy of NLMs decreases significantly. This paper further develop the NLMs-based denoising method to remove noise with less loss of image details. It is realized by embedding an optimal graph edge weights driven NLMs kernel into a multi-layer residual compensation framework. Unlike the patch similarity-based weights in the traditional NLMs filters, the edge weights derived from the optimal graph Laplacian regularization consider (1) the distance between the target pixel and the candidate pixel, (2) the local gradient and (3) the patch similarity. After defining the weights, the graph-based NLMs kernel is then put into a multi-layer framework. The corresponding primal and residual terms at each layer are finally fused with learned weights to recover the image. Experimental results show that our method is effective and robust, especially for piecewise smooth images. [ABSTRACT FROM AUTHOR]

Sparse representation models have been widely used in single-image super-resolution reconstruction. The construction of dictionaries is especially important in these models. Basically, approaches of dictionary construction in sparse representation can be divided into two categories: analytical and learning-based approaches. Analytical approaches are effective and fast, but they are unable to fit different types of data; learning-based approaches are adaptive, but their implementation takes a significant amount of time. In this study, an image super-resolution reconstruction algorithm based on double-sparsity dictionaries is proposed. The algorithm combines the efficiency of analytical approaches and adaptability of learning-based approaches. In addition to the sparsity prior, the non-local self-similarity prior is also considered in the algorithm. Non-local means filtering is used to be the constraints on regularized super-resolution reconstruction procedures, which improves the quality of super-resolution reconstruction results further, while the runtime of the algorithm is still acceptable. Experimental results demonstrate the advantage of the proposed algorithm. [ABSTRACT FROM AUTHOR]

• A voxel-based PVC method under the least squares framework (LS). • An anatomical-guided NLM (NLMA) can improve resolution and suppress image noise. • We use subtle usage of the accompanying MRI information to define the search space. • We evaluated this method using the BrainWeb phantom and in vivo human data. Positron emission tomography (PET) images tend to be significantly degraded by the partial volume effect (PVE) resulting from the limited spatial resolution of the reconstructed images. Our purpose is to propose a partial volume correction (PVC) method to tackle this issue. In the present work, we explore a voxel-based PVC method under the least squares framework (LS) employing anatomical non-local means (NLMA) regularization. The well-known non-local means (NLM) filter utilizes the high degree of information redundancy that typically exists in images, and is typically used to directly reduce image noise by replacing each voxel intensity with a weighted average of its non-local neighbors. Here we explore NLM as a regularization term within iterative-deconvolution model to perform PVC. Further, an anatomical-guided version of NLM was proposed that incorporates MRI information into NLM to improve resolution and suppress image noise. The proposed approach makes subtle usage of the accompanying MRI information to define a more appropriate search space within the prior model. To optimize the regularized LS objective function, we used the Gauss-Seidel (GS) algorithm with the one-step-late (OSL) technique. After the import of NLMA, the visual and quality results are all improved. With a visual check, we notice that NLMA reduce the noise compared to other PVC methods. This is also validated in bias-noise curve compared to non-MRI-guided PVC framework. We can see NLMA gives better bias-noise trade-off compared to other PVC methods. Our efforts were evaluated in the base of amyloid brain PET imaging using the BrainWeb phantom and in vivo human data. We also compared our method with other PVC methods. Overall, we demonstrated the value of introducing subtle MRI-guidance in the regularization process, the proposed NLMA method resulting in promising visual as well as quantitative performance improvements. [ABSTRACT FROM AUTHOR]

In this paper, we address the problem of denoising images obtained under low-light conditions for the Poisson shot noise model. Under such conditions, the variance stabilization transform (VST) is no longer applicable, so that the state-of-the-art algorithms which are proficient for the additive white Gaussian noise cannot be applied. We first introduce an oracular non-local algorithm and prove its convergence with the optimal rate of convergence under a Hölder regularity assumption for the underlying image, when the search window size is suitably chosen. We also prove that the convergence remains valid when the oracle function is estimated within a prescribed error range. We then define a realizable filter by a statistical estimation of the similarity function which determines the oracle weight. The convergence of the realizable filter is justified by proving that the estimator of the similarity function lies in the prescribed error range with high probability. The experiments show that under low-light conditions the proposed filter is competitive compared with the recent state-of-the-art algorithms. [ABSTRACT FROM AUTHOR]