Non-local Means Denoising Research Articles

RP squeeze U-SegNet model for lesion segmentation and optimization enabled ShuffleNet based multi-level severity diabetic retinopathy classification

ABSTRACT In Diabetic Retinopathy (DR), the retina is harmed due to the high blood pressure in small blood vessels. Manual screening is time-consuming, which can be overcome by using automated techniques. Hence, this paper proposed a new method for classifying the multi-level severity of DR. Initially, the input fundus image is pre-processed by Non-local means Denoising (NLMD). Then, lesion segmentation is carried out by the Recurrent Prototypical-squeeze U-SegNet (RP-squeeze U-SegNet). Next, feature extraction is effectuated to mine image-level features. DR is categorized as abnormal or normal by ShuffleNet and it is tuned by Fractional War Royale Optimization (FrWRO), and later, if DR is detected, severity classification is performed. Furthermore, the FrWRO-SqueezeNet obtained the maximum performance with sensitivity of 97%, accuracy of 93.8%, specificity of 95.1%, precision of 91.8%, and F-Measure of 94.3%. The devised scheme accurately visualizes abnormal regions in the fundus images. Also, it has the ability to identify the severity levels of DR effectively, which avoids the progression risk to vision loss and proliferative disease.

Parametric noise reduction and construction of performance degradation indexes for low-voltage switchgear appliances

Aiming at the problem that there is a lot of noise in the feature parameter data of low-voltage switchgear, which makes it difficult to extract useful information about the feature parameters and unable to construct the performance degradation index reasonably. This paper proposes a noise reduction method through Relative Entropy (KL) optimized Variational Mode Decomposition (VMD) combined with Non-Local Mean (NLM) and a method of constructing performance degradation indexes for low-voltage switching appliances through Convolutional Autoencoder (CAE). Firstly, the voltage and current signals are collected based on the full-life test platform of low-voltage switchgear, and the subset of feature parameters is extracted. Then, the feature parameters are decomposed into different components by relative entropy optimization VMD and the noise-dominant and effective components are selected based on the correlation. Secondly, NLM denoising is performed on the noise-dominated signal, and the noise-reduced signals are obtained by signal reconstruction. Finally, the construction of performance degradation indexes for multi-type switchgear is completed based on CAE. The final results show that the two noise reduction evaluation indicators including Noise Rejection Ratio (NRR) and trendiness (T), and the three performance degradation evaluation indexes including Goodness of Fit (R), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE) all show optimal results, indicating that this paper’s method can better complete the construction of noise reduction and performance degradation indicators for the feature parameters of low-voltage switchgear.

Multiplane HiLo microscopy with speckle illumination and non-local means denoising.

HiLo microscopy synthesizes an optically sectioned image from two images, one obtained with uniform and another with patterned illumination, such as laser speckle. Speckle-based HiLo has the advantage of being robust to aberrations but is susceptible to residual speckle noise that is difficult to control. We present a computational method to reduce this residual noise without undermining resolution. In addition, we improve the versatility of HiLo microscopy by enabling simultaneous multiplane imaging (here nine planes). Our goal is to perform fast, high-contrast, multiplane imaging with a conventional camera-based fluorescence microscope. Multiplane HiLo imaging is achieved with the use of a single camera and z-splitter prism. Speckle noise reduction is based on the application of a non-local means (NLM) denoising method to perform ensemble averaging of speckle grains. We demonstrate the capabilities of multiplane HiLo with NLM denoising both with synthesized data and by imaging cardiac and brain activity in zebrafish larvae at 40Hz frame rates. Multiplane HiLo microscopy aided by NLM denoising provides a simple tool for fast optically sectioned volumetric imaging that can be of general utility for fluorescence imaging applications.

Improved Preprocessing Strategy under Different Obscure Weather Conditions for Augmenting Automatic License Plate Recognition

Automatic license plate recognition (ALPR) systems are widely used for various applications, including traffic control, law enforcement, and toll collection. However, the performance of ALPR systems is often compromised in challenging weather and lighting conditions. This research aims to improve the effectiveness of ALPR systems in foggy, low-light, and rainy weather conditions using a hybrid preprocessing methodology. The research proposes the combination of dark channel prior (DCP), non-local means denoising (NMD) technique, and adaptive histogram equalization (AHE) algorithms in CIELAB color space. And used the Python programming language comparisons for SSIM and PSNR performance. The results showed that this hybrid approach is not merely robust to a variety of challenging conditions, including challenging weather and lighting conditions but significantly more accurate for existing ALPR systems.

Non-Local Means 잡음 제거와 데이터 증강을 이용한 YOLO 기반 객체 특징 탐색

When fruits are harvested in farms, most of them go through a manual sorting process and classify and distribute decomposed fruits. However, there is a limit to manually classifying large amounts in a situation where the number of workers is decreasing in farms. To solve this problem, it is important to divide normal and decomposed fruits in real time to minimize the proportion of manpower used in the screening process. We propose a method of YOLO based Object Features Detection Using Non-Local Means Denoising and Data Augmentation. The proposed method collects desired data through the Crawling method, and the preprocessing minimizes image noise through Non-Local Means Denoising. The built image dataset uses YOLOv3, an object detection algorithm, to distinguish and detect normal and decayed fruits. As a result of performance evaluation, object detection of YOLOv3 objects in a proposed method rather than detection results shows that Recall increases 10% performance and increases 9% in the remaining Recall and IoU. Therefore, the proposed method can increase screening efficiency by detecting decayed fruits well.

Low-Power Communication Signal Enhancement Method of Internet of Things Based on Nonlocal Mean Denoising

In order to improve the transmission effect of low-power communication signal of Internet of Things and compress the enhancement time of low-power communication signal, this paper designs a low-power communication signal enhancement method of Internet of Things based on nonlocal mean denoising. Firstly, the residual of one-dimensional communication layer is preprocessed by convolution core to obtain the residual of one-dimensional communication layer. Then, according to the two classification recognition methods, the noise reduction signal feature recognition of the low-power communication signal of the Internet of Things is realized, the nonlocal mean noise reduction algorithm is used to remove the low-power communication signal of the Internet of Things, and the weight value between similar blocks is calculated according to the European distance method. Finally, the low-power communication signal enhancement of the Internet of Things is realized by the nonlocal mean value denoising method. The experimental results show that the communication signal enhancement time overhead of this method is low, which is always less than 2.6 s. The lowest bit error rate after signal enhancement is about 1%, and the signal-to-noise ratio is up to 18 dB, which shows that this method can achieve signal enhancement.

Detection of Impurity Rate of Machine-Picked Cotton Based on Improved Canny Operator

Aiming at the real-time detection of the impurity rate in machine-picked cotton processing, a detection method for the impurity rate in machine-picked cotton was proposed based on an improved Canny operator. According to the characteristics of different saturations between cotton and impurities, the impurities were separated by extracting the image S channel. Due to problems existing in the traditional Canny operator’s edge detection, the Gaussian filter was replaced by employing mean filtering and nonlocal mean denoising, which could effectively remove the noise in the image. A YOLO V5 neural network was used to classify and identify the impurities after segmentation, and the densities of various impurities were measured. The volume–weight (V–W) model was established to solve the impurity rate based on mass. Compared with a single thread, the data processing time was shortened by 43.65%, and the frame rate was effectively improved by using multithread technology. By solving the average value of the impurity rate, the anti-interference performance of the algorithm was enhanced, and has the characteristics of real-time detection and stability. This method solved the problems of low speed, poor real-time detection, and ease of interference, and can be used to guide the cotton production process.

Image Noise Removal Method Based on Improved Nonlocal Mean Algorithm

Aiming at the problem of unclear images acquired in interactive systems, an improved image processing algorithm for nonlocal mean denoising is proposed. This algorithm combines the adaptive median filter algorithm with the traditional nonlocal mean algorithm, first adjusts the image window adaptively, selects the corresponding pixel weight, and then denoises the image, which can have a good filtering effect on the mixed noise. The experimental results show that, compared with the traditional nonlocal mean algorithm, the algorithm proposed in this paper has better results in the visual quality and peak signal-to-noise ratio (PSNR) of complex noise images.

Fuzzy Based Fast Non Local Mean Filter to Denoise Rician Noise

Abstract MRI images suffer from a type of multiplicative noise called Rician noise. The traditional smoothing filters are not effective to reduce this type of noise. Non local Means (NLM) filters with bias correction is a promising technique for such type of signal dependent noise. The NLM denoising is based on self-similarity and is computationally very expensive. Many researchers are working to improve its time complexity. In most cases there is a tradeoff between speed and quality of denoised image. This paper proposes a speed enhanced NLM method that use integral image and fuzzy Jacord similarity measure to find the self-similarity of sub windows. This helps to find the weights of similar pixel at a faster rate than the traditional NLM algorithm and more accurately than the existing fast NLM method. Consequently, these similar pixels are used to generate the noise free pixels. At the end the conventional bias subtraction method is used as post processing step. The proposed scheme is tested with the standard brain web data set with different noise levels and compared with existing Fast NLM techniques and basic NLM using Root mean square error (RMSE), peak signal noise ratio (PSNR), Structure similar index (SSIM), quality index and computational time parameters methods. The proposed method gives better result than existing Fast NLM technique with high density Rician noise in the image and is 20 times faster than traditional NLM.

Fuzzy based fast non local mean filter to denoise Rician noise

MRI images suffer from a type of multiplicative noise called Rician noise. The traditional smoothing filters are not effective to reduce this type of noise. Non local Means (NLM) filters with bias correction is a promising technique for such type of signal dependent noise. The NLM denoising is based on self-similarity and is computationally very expensive. Many researchers are working to improve its time complexity. In most cases there is a tradeoff between speed and quality of denoised image. This paper proposes a speed enhanced NLM method that use integral image and fuzzy Jacord similarity measure to find the self-similarity of sub windows. This helps to find the weights of similar pixel at a faster rate than the traditional NLM algorithm and more accurately than the existing fast NLM method. Consequently, these similar pixels are used to generate the noise free pixels. At the end the conventional bias subtraction method is used as post processing step. The proposed scheme is tested with the standard brain web data set with different noise levels and compared with existing Fast NLM techniques and basic NLM using Root mean square error (RMSE), peak signal noise ratio (PSNR), Structure similar index (SSIM), quality index and computational time parameters methods. The proposed method gives better result than existing Fast NLM technique with high density Rician noise in the image and is 20 times faster than traditional NLM.

Nonlocal Means Image Denoising With Minimum MSE-Based Decay Parameter Adaptation

Nonlocal means (NLM), a patch-based nonlocal recovery paradigm, has attracted much attention in recent decades. The decay parameter will greatly affect restoration performance of the NLM method. However, the existing NLM methods with decay parameter adaptation cannot determine this parameter effectively. To address this problem, we have proposed the minimum mean square error (MSE) based decay parameter adaptation method for the NLM denoising. In the proposed method, the globally optimal decay parameter is determined to produce the pre-denoised image based on the derived relation between the global minimum MSE and the decay parameter. Then, the pixel-wise MSE is estimated based on the pre-denoised result and the corresponding method noise. Finally, the optimal pixel-wise decay parameter is obtained by minimizing the pixel-wise MSE to produce the estimated restored image and the boosting strategy is implemented on this image to generate the final denoised result. Extensive simulations on the standard test images and real images corrupted with Gaussian noise and speckle noise demonstrate that the proposed method significantly outperforms some compared NLM methods in noise reduction and detail preservation and can provide better restoration performance than other state-of-the-art denoising methods in most cases in terms of objective metrics and human vision.

COGL: Coefficient Graph Laplacians for Optimized JPEG Image Decoding.

We address the problem of decoding joint photographic experts group (JPEG)-encoded images with less visual artifacts. We view the decoding task as an ill-posed inverse problem and find a regularized solution using a convex, graph Laplacian-regularized model. Since the resulting problem is non-smooth and entails non-local regularization, we use fast high-dimensional Gaussian filtering techniques with the proximal gradient descent method to solve our convex problem efficiently. Our patch-based "coefficient graph" is better suited than the traditional pixel-based ones for regularizing smooth non-stationary signals such as natural images and relates directly to classic non-local means de-noising of images. We also extend our graph along the temporal dimension to handle the decoding of M-JPEG-encoded video. Despite the minimalistic nature of our convex problem, it produces decoded images with similar quality to other more complex, state-of-the-art methods while being up to five times faster. We also expound on the relationship between our method and the classic ANCE method, reinterpreting ANCE from a graph-based regularization perspective.

Non local means denoising in photoacoustic imaging

Photoacoustic (PA) imaging has the ability to visualize human organs with high spatial resolution and high contrast. Like digital images, PA images are contaminated with random noise due to some parameters. The band-pass filter does not effectively remove the noise because noise is randomly distributed in the bandwidth frequency. We present noise removal method in PA images by using non local means denoising (NLMD) method. The NLMD can be used if there are similarities or redundancies in the image. PA images contain of blood vessel which repeating on the small patch. The method was tested on PA images of carbon nanotubes in micropipe, in vivo mice brain and in vivo mice ear. We estimated the suggested input parameters of NLMD, so it can be automatically applied after scanning the image in PA imaging system. Our results declared that the NLMD enhanced the image quality of PA images.

Artifact reduction for separable nonlocal means

It was recently demonstrated [J. Electron. Imaging, 25(2), 2016] that one can perform fast non-local means (NLM) denoising of one-dimensional signals using a method called lifting. The cost of lifting is independent of the patch length, which dramatically reduces the run-time for large patches. Unfortunately, it is difficult to directly extend lifting for non-local means denoising of images. To bypass this, the authors proposed a separable approximation in which the image rows and columns are filtered using lifting. The overall algorithm is significantly faster than NLM, and the results are comparable in terms of PSNR. However, the separable processing often produces vertical and horizontal stripes in the image. This problem was previously addressed by using a bilateral filter-based post-smoothing, which was effective in removing some of the stripes. In this letter, we demonstrate that stripes can be mitigated in the first place simply by involving the neighboring rows (or columns) in the filtering. In other words, we use a two-dimensional search (similar to NLM), while still using one-dimensional patches (as in the previous proposal). The novelty is in the observation that one can use lifting for performing two-dimensional searches. The proposed approach produces artifact-free images, whose quality and PSNR are comparable to NLM, while being significantly faster.

Periodicity-based nonlocal-means denoising method for electrocardiography in low SNR non-white noisy conditions

Abstract Nonlocal means (NLM) denoising, originally developed for non-neighborhood image filtering, might be appropriate for denoising electrocardiography (ECG). It was applied to ECG signals and achieved results comparable to those of other state-of-the-art filters. This study proposed periodic NLM filtering (pNLM) for ECG and tested it in various noise environments. To increase the original NLM denoising performance for ECG, pNLM search windows were selected based on ECG periodicity, reducing dissimilar patch effects and leading to better denoising performance. The algorithm was evaluated using the MIT-BIH arrhythmia database and quantitative metrics, such as signal-to-noise ratio (SNR) improvement, mean squared error (MSE), and percent root mean square difference (PRD). Experimental results showed that this novel denoising method increased denoising performance compared to the NLM method by 23.8%, 28.8% and 97.9% for white, pink, and electromyogram (EMG) noise, respectively, especially for low SNR input. In summary, the pNLM algorithm is effective for denoising three types of ECG noise.

Looking for the Signal: A guide to iterative noise and artefact removal in X-ray tomographic reconstructions of porous geomaterials

X-ray micro- and nanotomography has evolved into a quantitative analysis tool rather than a mere qualitative visualization technique for the study of porous natural materials. Tomographic reconstructions are subject to noise that has to be handled by image filters prior to quantitative analysis. Typically, denoising filters are designed to handle random noise, such as Gaussian or Poisson noise. In tomographic reconstructions, noise has been projected from Radon space to Euclidean space, i.e. post reconstruction noise cannot be expected to be random but to be correlated. Reconstruction artefacts, such as streak or ring artefacts, aggravate the filtering process so algorithms performing well with random noise are not guaranteed to provide satisfactory results for X-ray tomography reconstructions. With sufficient image resolution, the crystalline origin of most geomaterials results in tomography images of objects that are untextured. We developed a denoising framework for these kinds of samples that combines a noise level estimate with iterative nonlocal means denoising. This allows splitting the denoising task into several weak denoising subtasks where the later filtering steps provide a controlled level of texture removal. We describe a hands-on explanation for the use of this iterative denoising approach and the validity and quality of the image enhancement filter was evaluated in a benchmarking experiment with noise footprints of a varying level of correlation and residual artefacts. They were extracted from real tomography reconstructions. We found that our denoising solutions were superior to other denoising algorithms, over a broad range of contrast-to-noise ratios on artificial piecewise constant signals.

혼합 norm 기반의 가중치 함수를 이용한 평균 노이즈 제거 기법

본 논문에서는 혼합 norm을 이용한 가중치 함수 기반의 비국부 평균 노이즈 제거 방식을 제안한다. 비국부 평균 노이즈 제거 방식에서 중심 패치와 참조 패치의 오차에 대한 신뢰도는 노이즈 양 및 국부 활동성에 의존적인 특성을 갖고 있다. 본 논문에서는 혼합 norm 기반의 새로운 가중치 함수를 제안하고, 혼합 norm의 차수를 노이즈 정도 및 중심 패치의 국부 활동성에 의해 적응적으로 결정하여 비국부 평균 노이즈 제거 방식의 성능을 개선하고자 하였다. 실험 결과를 통해 기존의 비국부 평균 노이즈 제거 방식과 비교하여 제안 방식의 정량적 및 정성적 성능의 우수성을 확인할 수 있었다. 더불어, 제안 방식은 표준 유클리드 norm 기반의 다른 형태의 비국부 평균 노이즈 방식의 성능을 개선할 수 있는 능력이 있음을 확인할 수 있었다. This paper presents a non-local means (NLM) denoising algorithm based on a new weighting function using a mixed norm. The fidelity of the difference between an anchor patch and the reference patch in the NLM denoising depends on noise level and local activity. This paper introduces a new weighting function based on a mixed norm type of which the order is determined by noise level and local activity of an anchor patch, so that the performance of the NLM denoising can be enhanced. Experimental results demonstrate the objective and subjective capability of the proposed algorithm. In addition, it was verified that the proposed algorithm can be used to improve the performance of the other <TEX>$l_2$</TEX> norm based non-local means denoising algorithms

Fast non‐local means with size‐adaptive search window

In this study, the authors present a fast non-local means (NLM) image denoising algorithm with size-adaptive search window. On the basis of the edge gradient and direction of the noisy image, the proposed fast scheme divides all the pixels into significant edge, moderate edge or non-edge region, and the size of the search window for most pixels which belong to non-edge region can be reduced. The proposed fast scheme also adopts different strategies to pre-select similar patches in the search window for efficient NLM denoising. Experimental results show that compared with the standard NLM method, the proposed fast scheme achieves a substantial reduction in computational cost and improvement in the denoising performance, both in terms of visual quality and numerical results.

Rolling element bearing fault diagnosis based on non‐local means de‐noising and empirical mode decomposition

The presence of faults in the bearings of rotating machinery is usually observed with impulses in the vibration signals. However, the vibration signals are generally non-stationary and usually contaminated by noise because of the compounded background noise present in the measuring device and the effect of interference from other machine elements. Therefore in order to enhance monitoring condition, the vibration signal needs to be properly de-noised before analysis. In this study, a novel fault diagnosis method for rolling element bearings is proposed based on a hybrid technique of non-local means (NLM) de-noising and empirical mode decomposition (EMD). An NLM which removes the noise with minimal signal distortion is first employed to eliminate or at least reduce the background noise present in the measuring device. This de-noised signal is then decomposed into a finite number of stationary intrinsic mode functions (IMF) to extract the impulsive fault features from the effect of interferences from other machine elements. Finally, envelope analyses are performed for IMFs to allow for easier detection of such characteristic fault frequencies. The results of simulated and real bearing vibration signal analyses show that the hybrid feature extraction technique of NLM de-noising, EMD and envelope analyses successfully extract impulsive features from noise signals.

Using fMRI non-local means denoising to uncover activation in sub-cortical structures at 1.5 T for guided HARDI tractography.

In recent years, there has been ever-increasing interest in combining functional magnetic resonance imaging (fMRI) and diffusion magnetic resonance imaging (dMRI) for better understanding the link between cortical activity and connectivity, respectively. However, it is challenging to detect and validate fMRI activity in key sub-cortical areas such as the thalamus, given that they are prone to susceptibility artifacts due to the partial volume effects (PVE) of surrounding tissues (GM/WM interface). This is especially true on relatively low-field clinical MR systems (e.g., 1.5 T). We propose to overcome this limitation by using a spatial denoising technique used in structural MRI and more recently in diffusion MRI called non-local means (NLM) denoising, which uses a patch-based approach to suppress the noise locally. To test this, we measured fMRI in 20 healthy subjects performing three block-based tasks : eyes-open closed (EOC) and left/right finger tapping (FTL, FTR). Overall, we found that NLM yielded more thalamic activity compared to traditional denoising methods. In order to validate our pipeline, we also investigated known structural connectivity going through the thalamus using HARDI tractography: the optic radiations, related to the EOC task, and the cortico-spinal tract (CST) for FTL and FTR. To do so, we reconstructed the tracts using functionally based thalamic and cortical ROIs to initiates seeds of tractography in a two-level coarse-to-fine fashion. We applied this method at the single subject level, which allowed us to see the structural connections underlying fMRI thalamic activity. In summary, we propose a new fMRI processing pipeline which uses a recent spatial denoising technique (NLM) to successfully detect sub-cortical activity which was validated using an advanced dMRI seeding strategy in single subjects at 1.5 T.