Validity of non-local mean filter and novel denoising method

Background Image denoising is an important topic in the digital image processing field.This study theoretically investigates the validity of the classical nonlocal mean filter(NLM)for removing Gaussian noise from a novel statistical perspective.Method By considering the restored image as an estimator of the clear image from a statistical perspective,we gradually analyze the unbiasedness and effectiveness of the restored value obtained by the NLM filter.Subsequently,we propose an improved NLM algorithm called the clustering-based NLM filter that is derived from the conditions obtained through the theoretical analysis.The proposed filter attempts to restore an ideal value using the approximately constant intensities obtained by the image clustering process.In this study,we adopt a mixed probability model on a prefiltered image to generate an estimator of the ideal clustered components.Result The experiment yields improved peak signal-to-noise ratio values and visual results upon the removal of Gaussian noise.Conclusion However,the considerable practical performance of our filter demonstrates that our method is theoretically acceptable as it can effectively estimate ideal images.

基于结构张量的Non-Local Means去噪算法研究

非局部平均是当前一种新兴而有效的图像去噪方法。为了能充分利用数字图像局部几何结构的自相似性,同时由于结构张量可有效刻画数字图像的局部几何结构特征,进而提出了基于结构张量相似性度量的非局部平均去噪算法。实验结果验证了该算法抑制噪声的有效性,同时能很好地保持边缘等细节特征,峰值信噪比得到有效提高。

基于Non-Local means滤波的雾天降质图像恢复算法

针对目前去雾算法易导致边缘晕环效应、边缘轮廓及景物特征比较模糊问题,提出了一种景深等先验信息未知条件下基于Non-Local means滤波的雾天降质图像恢复算法。首先,根据大气散射模型将经典的场景深度估计转化为大气面纱以及天空亮度估计,避免难求的场景深度图;然后,对雾天降质图像进行雾气平均化预处理,经过预处理图像平均亮度变小;其次,依据大气面纱的边缘跟雾天图像的低频具有大的相似性,采用Non-Localmeans滤波算法估计大气面纱模型;最后,为了使恢复图像的亮度跟色度都更加接近晴天图像,进行防止对比度放大的平滑与色度调整处理。通过与已有实验结果对比表明,提出的算法可以获得更精确的大气面纱,恢复图像不但边缘轮廓及景物特征都比较清楚,而且可有效抑制边缘晕环效应。

Fast Non-Local Means Algorithm Based on Krawtchouk Moments

Non-local means(NLM)method is a state-of-the-art denoising algorithm, which replaces each pixel with a weighted average of all the pixels in the image. However, the huge computational complexity makes it impractical for real applications. Thus, a fast non-local means algorithm based on Krawtchouk moments is proposed to improve the denoising performance and reduce the computing time. Krawtchouk moments of each image patch are calculated and used in the subsequent similarity measure in order to perform a weighted averaging. Instead of computing the Euclidean distance of two image patches, the similarity measure is obtained by low-order Krawtchouk moments, which can reduce a lot of computational complexity. Since Krawtchouk moments can extract local features and have a good antinoise ability, they can classify the useful information out of noise and provide an accurate similarity measure. Detailed experiments demonstrate that the proposed method outperforms the original NLM method and other moment-based methods according to a comprehensive consideration on subjective visual quality, method noise, peak signal to noise ratio(PSNR), structural similarity(SSIM) index and computing time. Most importantly, the proposed method is around 35 times faster than the original NLM method.

Improved Non-Local Means Algorithm for Image Denoising

Image denoising technology is one of the forelands in the field of computer graphic and computer vision. Non-local means method is one of the great performing methods which arouse tremendous research. In this paper, an improved weighted non-local means algorithm for image denoising is proposed. The non-local means denoising method replaces each pixel by the weighted average of pixels with the surrounding neighborhoods. The proposed method evaluates on testing images with various levels noise. Experimental results show that the algorithm improves the denoising performance.

Two Modifications of Weight Calculation of the Non-Local Means Denoising Method

The non-local means (NLM) denoising method replaces each pixel by the weighted average of pixels with the sur-rounding neighborhoods. In this paper we employ a cosine weighting function instead of the original exponential func-tion to improve the efficiency of the NLM denoising method. The cosine function outperforms in the high level noise more than low level noise. To increase the performance more in the low level noise we calculate the neighborhood si-milarity weights in a lower-dimensional subspace using singular value decomposition (SVD). Experimental compari-sons between the proposed modifications against the original NLM algorithm demonstrate its superior denoising per-formance in terms of peak signal to noise ratio (PSNR) and histogram, using various test images corrupted by additive white Gaussian noise (AWGN).

Long radial coherence of electron temperature fluctuations in non-local transport in HL-2A plasmas

The dynamics of long-wavelength(kθ<1.4 cm^(-1)),broadband(20 kHz–200 kHz)electron temperature fluctuations(Te/Te)of plasmas in gas-puff experiments are observed for the first time in HL-2A tokamak.In a relatively low density(ne(0)■0.91×10^(19)m^(-3)–1.20×10^(19)m^(-3))scenario,after gas-puffing the core temperature increases and the edge temperature drops.On the contrary,temperature fluctuation drops at the core and increases at the edge.Analyses show the non-local emergence is accompanied with a long radial coherent length of turbulent fluctuations.While in a higher density(ne(0)?1.83×10^(19)m^(-3)–2.02×10^(19)m^(-3))scenario,the phenomena are not observed.Furthermore,compelling evidence indicates that E×B shear serves as a substantial contributor to this extensive radial interaction.This finding offers a direct explanatory link to the intriguing core-heating phenomenon witnessed within the realm of non-local transport.

Fresnel Equations Derived Using a Non-Local Hidden-Variable Particle Theory

Problem: The Fresnel equations describe the proportions of reflected and transmitted light from a surface, and are conventionally derived from wave theory continuum mechanics. Particle-based derivations of the Fresnel equations appear not to exist. Approach: The objective of this work was to derive the basic optical laws from first principles from a particle basis. The particle model used was the Cordus theory, a type of non-local hidden-variable (NLHV) theory that predicts specific substructures to the photon and other particles. Findings: The theory explains the origin of the orthogonal electrostatic and magnetic fields, and re-derives the refraction and reflection laws including Snell’s law and critical angle, and the Fresnel equations for s and p-polarisation. These formulations are identical to those produced by electromagnetic wave theory. Contribution: The work provides a comprehensive derivation and physical explanation of the basic optical laws, which appears not to have previously been shown from a particle basis. Implications: The primary implications are for suggesting routes for the theoretical advancement of fundamental physics. The Cordus NLHV particle theory explains optical phenomena, yet it also explains other physical phenomena including some otherwise only accessible through quantum mechanics (such as the electron spin g-factor) and general relativity (including the Lorentz and relativistic Doppler). It also provides solutions for phenomena of unknown causation, such as asymmetrical baryogenesis, unification of the interactions, and reasons for nuclide stability/instability. Consequently, the implication is that NLHV theories have the potential to represent a deeper physics that may underpin and unify quantum mechanics, general relativity, and wave theory.

A Robust and Fast Non-Local Means Algorithm for Image Denoising

In the paper, we propose a robust and fast image denoising method. The approach integrates both Non- Local means algorithm and Laplacian Pyramid. Given an image to be denoised, we first decompose it into Laplacian pyramid. Exploiting the redundancy property of Laplacian pyramid, we then perform non-local means on every level image of Laplacian pyramid. Essentially, we use the similarity of image features in Laplacian pyramid to act as weight to denoise image. Since the features extracted in Laplacian pyramid are localized in spatial position and scale, they are much more able to describe image, and computing the similarity between them is more reasonable and more robust. Also, based on the efficient Summed Square Image （SSI） scheme and Fast Fourier Transform （FFT）, we present an accelerating algorithm to break the bottleneck of non-local means algorithm - similarity computation of compare windows. After speedup, our algorithm is fifty times faster than original non-local means algorithm. Experiments demonstrated the effectiveness of our algorithm.

The Algorithms about Fast Non-local Means Based Image Denoising

Image denoising is still a challenge of image processing. Buades et al. proposed a nonlocal means （NL-means） approach. This method had a remarkable denoising results at high expense of computational cost. In this paper, We compared several fast non-local means methods, and proposed a new fast algorithm. Numerical experiments showed that our algorithm considerably reduced the computational cost, and obtained visually pleasant images.

A Two-Step Regularization Framework for Non-Local Means

As an effective patch-based denoising method, non-local means （NLM） method achieves favorable denoising performance over its local counterparts and has drawn wide attention in image processing community. The in, plementation of NLM can formally be decomposed into two sequential steps, i.e., computing the weights and using the weights to compute the weighted means. In the first step, the weights can be obtained by solving a regularized optimization. And in the second step, the means can be obtained by solving a weighted least squares problem. Motivated by such observations, we establish a two-step regularization framework for NLM in this paper. Meanwhile, using the fl-amework, we reinterpret several non-local filters in the unified view. Further, taking the framework as a design platform, we develop a novel non-local median filter for removing salt-pepper noise with encouraging experimental results.

Mean Shift跟踪算法创新实验项目设计

视频跟踪算法是计算机视觉实践课程中比较受关注的实验项目。针对突变情况下传统Mean Shift跟踪算法无法实时准确跟踪的问题,设计了基于模板更新和线性预估的Mean Shift跟踪算法创新实验项目。在模板更新策略下,引入背景模板,通过将原目标模板和背景模板与设定的阈值进行比较来对干扰因素进行判定,当干扰因素判定目标受到遮挡时,引入线性预估方程进行目标位置预测,有效解决目标在遮挡情况下跟踪丢失的问题。通过对测试视频的跟踪效果和性能进行对比分析,验证了算法在突变情况下相较于传统算法具有更好的抗干扰能力。以算法创新设计为核心,通过开放性创新实验项目的选题、设计、答辩、反馈的闭环实验过程,有效提高了学生算法创新设计能力。

加入跳跃连接的深度嵌入K-means聚类

现有的深度聚类算法大多采用对称的自编码器来提取高维数据的低维特征,但随着自编码器训练次数的不断增加,数据的低维特征空间在一定程度上发生了扭曲,这样得到的数据低维特征空间无法反映原始数据空间中潜在的聚类结构信息.为了解决上述问题,本文提出了一种新的深度嵌入K-means算法(SDEKC).首先,在低维特征提取阶段,在对称的卷积自编码器中相对应的编码器与解码器之间以一定的权重加入两个跳跃连接,以减弱解码器对编码器的编码要求同时突出卷积自编码器的编码能力,这样可以更好地保留原始数据空间中蕴含的聚类结构信息;其次,在聚类阶段,通过一个标准正交变换矩阵将低维数据空间转换为一个新的揭示聚类结构信息的空间;最后,本文以端到端的方式采用贪婪算法迭代优化数据的低维表示及其聚类,在6个真实数据集上验证了本文提出新算法的有效性.

基于蚁群算法的三支k-means聚类算法

在聚类分析中,三支k-means聚类算法较具有较强的处理边界不确定数据的能力,但仍然存在对初始聚类中心敏感的问题.通过将蚁群算法和三支k-means聚类算法相结合,给出了一种基于蚁群算法的三支k-means聚类算法来解决这一问题.利用蚁群算法中随机概率选择策略和信息素的正负反馈机制,动态调整权重的方法,对三支k-means聚类算法进行优化.在UCI数据集上实验证明,该方法对聚类结果的性能指标有所提高.

基于K-Means聚类与熵权TOPSIS法的岩石可爆性评价研究

露天矿山的爆破块度分布,直接影响到后续的采装、运输和破碎工作。为了控制石墨矿山不同区域爆破块度分布,基于K-means无监督聚类学习法与熵权TOPSIS评价法建立了一种新的岩石可爆性评价模型,选取岩石密度、动力能量耗散率、动态抗压强度、平均应变率、脆性指数作为评价指标,通过熵权计算,发现岩石破碎程度受脆性指数影响最大,受平均应变率影响最小。将此模型应用于实际石墨矿山,可爆性分为10个等级,统计不同分级下的岩石平均破碎粒径,发现可爆性分级等级越高平均粒径越大,有明显的分级特征,验证了模型的有效性。从爆破石墨矿石岩体类型看,岩石可爆性从易到难排序为:片岩、片麻岩、变粒岩、混合岩。结合石墨矿石微观观测结果分析可知:岩性从片岩向混合岩转变,岩石内部石墨晶质呈下降趋势,石墨矿石可爆性等级也随之越来越高。岩石密度、能量耗散率、动态抗压强度之间呈线性正相关,岩石可爆性与平均应变率、脆性指数存在负相关性。研究成果为矿山矿岩可爆性评价提供了一条新思路,对露天矿山爆破块度优化具有一定的理论和实践指导意义。

High performance“non-local”generic face reconstruction model using the lightweight Speckle-Transformer(SpT)UNet

Significant progress has been made in computational imaging(CI),in which deep convolutional neural networks(CNNs)have demonstrated that sparse speckle patterns can be reconstructed.However,due to the limited“local”kernel size of the convolutional operator,for the spatially dense patterns,such as the generic face images,the performance of CNNs is limited.Here,we propose a“non-local”model,termed the Speckle-Transformer(SpT)UNet,for speckle feature extraction of generic face images.It is worth noting that the lightweight SpT UNet reveals a high efficiency and strong comparative performance with Pearson Correlation Coefficient(PCC),and structural similarity measure(SSIM)exceeding 0.989,and 0.950,respectively.

光伏波动平抑下改进K-means的电池储能动态分组控制策略

针对电池储能系统(battery energy storage system,BESS)进行光伏波动平抑时寿命损耗高及荷电状态(state of charge,SOC)一致性差的问题,提出了光伏波动平抑下改进K-means的BESS动态分组控制策略。首先,采用最小最大调度方法获取光伏并网指令。其次,设计了改进侏儒猫鼬优化算法(improved dwarf mongoose optimizer,IDMO),并利用它对传统K-means聚类算法进行改进,加快了聚类速度。接着,制定了电池单元动态分组原则,并根据电池单元SOC利用改进K-means将其分为3个电池组。然后,设计了基于充放电函数的电池单元SOC一致性功率分配方法,并据此提出BESS双层功率分配策略,上层确定电池组充放电顺序及指令,下层计算电池单元充放电指令。对所提策略进行仿真验证,结果表明,所设计的IDMO具有更高的寻优精度及更快的寻优速度。所提BESS平抑光伏波动策略在有效平抑波动的同时,降低了BESS运行寿命损耗并提高了电池单元SOC的均衡性。

基于特征分箱和K-Means算法的用户行为分析方法

针对网购用户所产生的购物行为进行分析,首先通过数据处理构建客户关系管理模型(RFM模型),在此模型的基础上采用特征分箱法和K-Means聚类两种方法对用户进行细分,并对2种模型结果进行比较分析,讨论二者的差异性和具体的应用范围和意义.其中,基于特征分箱法的RFM模型将变量转化到相似的尺度上并将变量离散化,使得用户分类标签更加清晰,也可依据各类标签分类出不同类型的用户.K-Means算法通过轮廓系数评估聚类算法质量以至于选取最优K值.本文实验分析结果可为运营商提供更加可靠直观的数据,使得运营商可以根据不同用户的不同行为进行市场细分,进而进行精准营销和服务设置.

基于改进K-means与机器视觉的档案数据分析技术

为了提升医疗信息系统对健康档案数据的分析效率,文中采用图像采集、降噪、配准与差分等技术提取医疗图像信息,进而有效提升信息系统的数据采集效率。同时还对传统的K-means算法加以改进,并提出了一种基于类间、类内距离的聚类初始化评价指标体系(BWP),将其应用于采集到的档案数据中,以实现快速的聚类分析。将所提算法在CUDA计算平台上进行了实现,测试结果表明,该方法的聚类精度和运行效率较现有算法均有显著提升。此外,改进后K-means算法的正确聚类样本数量占比提升了4.88%,高于现有的主流指标体系,且当聚类数k的取值为16或32时,运行时间大幅降低。

基于K-means聚类和特征空间增强的噪声标签深度学习算法

深度学习中神经网络的性能依赖于高质量的样本,然而噪声标签会降低网络的分类准确率。为降低噪声标签对网络性能的影响,噪声标签学习算法被提出。该算法首先将训练样本集划分成干净样本集和噪声样本集,然后使用半监督学习算法对噪声样本集赋予伪标签。然而,错误的伪标签以及训练样本数量不足的问题仍然限制着噪声标签学习算法性能的提升。为解决上述问题,提出基于K-means聚类和特征空间增强的噪声标签深度学习算法。首先,该算法利用K-means聚类算法对干净样本集进行标签聚类,并根据噪声样本集与聚类中心的距离大小筛选出难以分类的噪声样本,以提高训练样本的质量;其次,使用mixup算法扩充干净样本集和噪声样本集,以增加训练样本的数量;最后,采用特征空间增强算法抑制mixup算法新生成的噪声样本,从而提高网络的分类准确率。并在CIFAR10、CIFAR100、MNIST和ANIMAL-10共4个数据集上试验验证了该算法的有效性。