A Second-Order Image Denoising Model for Contrast Preservation

In this work,we propose a second-order model for image denoising by employing a novel potential function recently developed in Zhu(J Sci Comput 88:46,2021)for the design of a regularization term.Due to this new second-order derivative based regularizer,the model is able to alleviate the staircase effect and preserve image contrast.The augmented Lagrangian method(ALM)is utilized to minimize the associated functional and convergence analysis is established for the proposed algorithm.Numerical experiments are presented to demonstrate the features of the proposed model.

BeFOI: A Novel Method Based on Conditional Diffusion Model for Medical Image Denoising

The progress in medical imaging technology highlights the importance of image quality for effective diagnosis and treatment.Yet,noise during capture and transmission can compromise image accuracy and reliability,complicating clinical decisions.The rising interest in diffusion models has led to their exploration of denoising images.We present Be-FOI(Better Fluoro Images),a weakly supervised model that uses cine images to denoise fluoroscopic images,both DR types.Trained through precise noise estimation and simulation,BeFOI employs Markov chains to denoise using only the fluoroscopic image as guidance.Our tests show that BeFOI outperforms other methods,reducing noise and enhancing clar-ity and diagnostic utility,making it an effective post-processing tool for medical images.

Image Denoising Combining the P-M Model and the LLT Model

In this paper, we present a noise removal technique by combining the P-M model with the LLT model. The combined technique takes full use of the advantage of both filters which is able to preserve edges and simultaneously overcomes the staircase effect. We use a weighting function in our model, and compare this model with the P-M model as well as other fourth-order functional both in theory and numerical experiment.

Image denoising using statistical model based on quaternion wavelet domain

Image denoising is the basic problem of image processing. Quaternion wavelet transform is a new kind of multiresolution analysis tools. Image via quaternion wavelet transform, wavelet coefficients both in intrascale and in interscale have certain correla- tions. First, according to the correlation of quaternion wavelet coefficients in interscale, non-Ganssian distribution model is used to model its correlations, and the coefficients are divided into important and unimportance coefficients. Then we use the non-Gaussian distribution model to model the important coefficients and its adjacent coefficients, and utilize the MAP method estimate original image wavelet coefficients from noisy coefficients, so as to achieve the purpose of denoising. Experimental results show that our al- gorithm outperforms the other classical algorithms in peak signal-to-noise ratio and visual quality.

Image Denoising Based on the Asymmetric Gaussian Mixture Model

In recent years,image restoration has become a huge subject,and finite hybrid model has been widely used in image denoising because of its easy modeling and strong explanatory results.The gaussian mixture model is the most common one.The existing image denoising methods usually assume that each component of the natural image is subject to the gaussian mixture model(GMM).However,this approach is not entirely reasonable.It is well known that most natural images are complex and their distribution is not entirely gaussian.As a result,there are still many problems that GMM cannot solve.This paper tries to improve the finite mixture model and introduces the asymmetric gaussian mixture model into it.Since the asymmetric gaussian mixture model can simulate the asymmetric distribution on the basis of the gaussian mixture model,it is more consistent with the natural image data,so the denoising effect of the natural complex image is better.We carried out image denoising experiments under different noise scales and types,and found that the asymmetric gaussian mixture model has better denoising effect and performance.

A Sobel-TV Based Hybrid Model for Robust Image Denoising

The traditional Total-Variation algorithm has a good result to de-noise for noise image of small scale details, but it easily losses the details for the image with rich texture and tiny boundary. In order to solve this problem, this paper proposes a Sobel-TV model algorithm for image denoising. It uses TV model to de-noise and uses Sobel algorithm to control smoothness of image, which not only efficiently removes image noise but also simultaneously retail information, such as edge and texture. The experiments demonstrate that the proposed algorithm is simple, practical and generates better SNR, which is an important value to preprocess image.

Seismic data denoising under the morphological component analysis framework combined with adaptive K-SVD and wave atoms dictionary

Many different effective reflection information are often contaminated by exterior and random noise which concealed in the seismic data.Traditional single or fixed transform is not suit for exploiting their complicated characteristics and attenuating the noise.Recent years,a novel method so-called morphological component analysis(MCA)is put forward to separate different geometrical components by amalgamating several irrelevance transforms.According to study the local singular and smooth linear components characteristics of seismic data,we propose a method of suppressing noise by integrating with the advantages of adaptive K-singular value decomposition(K-SVD)and wave atom dictionaries to depict the morphological features diversity of seismic signals.Numerical results indicate that our method can dramatically suppress the undesired noises,preserve the information of geologic body and geological structure and improve the signal-to-noise ratio of the data.We also demonstrate the superior performance of this approach by comparing with other novel dictionaries such as discrete cosine transform(DCT),undecimated discrete wavelet transform(UDWT),or curvelet transform,etc.This algorithm provides new ideas for data processing to advance quality and signal-to-noise ratio of seismic data.

A fast-convergence POCS seismic denoising and reconstruction method

The efficiency, precision, and denoising capabilities of reconstruction algorithms are critical to seismic data processing. Based on the Fourier-domain projection onto convex sets （POCS） algorithm, we propose an inversely proportional threshold model that defines the optimum threshold, in which the descent rate is larger than in the exponential threshold in the large-coefficient section and slower than in the exponential threshold in the small-coefficient section. Thus, the computation efficiency of the POCS seismic reconstruction greatly improves without affecting the reconstructed precision of weak reflections. To improve the flexibility of the inversely proportional threshold, we obtain the optimal threshold by using an adjustable dependent variable in the denominator of the inversely proportional threshold model. For random noise attenuation by completing the missing traces in seismic data reconstruction, we present a weighted reinsertion strategy based on the data-driven model that can be obtained by using the percentage of the data-driven threshold in each iteration in the threshold section. We apply the proposed POCS reconstruction method to 3D synthetic and field data. The results suggest that the inversely proportional threshold model improves the computational efficiency and precision compared with the traditional threshold models; furthermore, the proposed reinserting weight strategy increases the SNR of the reconstructed data.

Iterative regularization method for image denoising with adaptive scale parameter

In order to decrease the sensitivity of the constant scale parameter, adaptively optimize the scale parameter in the iteration regularization model （IRM） and attain a desirable level of applicability for image denoising, a novel IRM with the adaptive scale parameter is proposed. First, the classic regularization item is modified and the equation of the adaptive scale parameter is deduced. Then, the initial value of the varying scale parameter is obtained by the trend of the number of iterations and the scale parameter sequence vectors. Finally, the novel iterative regularization method is used for image denoising. Numerical experiments show that compared with the IRM with the constant scale parameter, the proposed method with the varying scale parameter can not only reduce the number of iterations when the scale parameter becomes smaller, but also efficiently remove noise when the scale parameter becomes bigger and well preserve the details of images.

TendiffPure:a convolutional tensor-train denoising diffusion model for purification

Diffusion models are effective purification methods,where the noises or adversarial attacks are removed using generative approaches before pre-existing classifiers conducting classification tasks.However,the efficiency of diffusion models is still a concern,and existing solutions are based on knowledge distillation which can jeopardize the generation quality because of the small number of generation steps.Hence,we propose TendiffPure as a tensorized and compressed diffusion model for purification.Unlike the knowledge distillation methods,we directly compress U-Nets as backbones of diffusion models using tensor-train decomposition,which reduces the number of parameters and captures more spatial information in multi-dimensional data such as images.The space complexity is reduced from O(N^(2))to O(NR^(2))with R≤4 as the tensor-train rank and N as the number of channels.Experimental results show that TendiffPure can more efficiently obtain high-quality purification results and outperforms the baseline purification methods on CIFAR-10,Fashion-MNIST,and MNIST datasets for two noises and one adversarial attack.

Chaotic signal denoising algorithm based on sparse decomposition

Denoising of chaotic signal is a challenge work due to its wide-band and noise-like characteristics.The algorithm should make the denoised signal have a high signal to noise ratio and retain the chaotic characteristics.We propose a denoising method of chaotic signals based on sparse decomposition and K-singular value decomposition(K-SVD)optimization.The observed signal is divided into segments and decomposed sparsely.The over-complete atomic library is constructed according to the differential equation of chaotic signals.The orthogonal matching pursuit algorithm is used to search the optimal matching atom.The atoms and coefficients are further processed to obtain the globally optimal atoms and coefficients by K-SVD.The simulation results show that the denoised signals have a higher signal to noise ratio and better preserve the chaotic characteristics.

Hybrid denoising-jittering data processing approach to enhance sediment load prediction of muddy rivers

Successful modeling of hydroenvironmental processes widely relies on quantity and quality of accessible data,and noisy data can affect the modeling performance.On the other hand in training phase of any Artificial Intelligence(AI) based model,each training data set is usually a limited sample of possible patterns of the process and hence,might not show the behavior of whole population.Accordingly,in the present paper,wavelet-based denoising method was used to smooth hydrological time series.Thereafter,small normally distributed noises with the mean of zero and various standard deviations were generated and added to the smooth time series to form different denoised-jittered data sets.Finally,the obtained pre-processed data were imposed into Artificial Neural Network(ANN) and Adaptive Neuro-Fuzzy Inference System(ANFIS)models for daily runoff-sediment modeling of the Minnesota River.To evaluate the modeling performance,the outcomes were compared with results of multi linear regression(MLR) and Auto Regressive Integrated Moving Average(ARIMA)models.The comparison showed that the proposed data processing approach which serves both denoising and jittering techniques could enhance the performance of ANN and ANFIS based runoffsediment modeling of the case study up to 34%and 25%in the verification phase,respectively.

实时VR场景下多尺度数字全息成像仿真

在数字全息成像的过程中,受光电传感器件结构参量、光照强度以及传输方向等因素的影响,导致全息图成像的亮度、分辨率降低。为了提高多尺度数字全息成像的真实度,提出了实时VR场景下多尺度数字全息成像方法。通过定义图像像素点分布的概率函数,计算多尺度数字图像像素的显著性,根据图像像素特征的分布方差,提取出多尺度数字图像像素点。利用小波变换系数识别多尺度数字图像的噪声,基于图像噪声信号的方差分布,精细化处理多尺度数字图像,结合小波反变换,构建多尺度数字图像去噪模型,得到去噪后的图像。根据多尺度数字全息成像模型,得到图像增强的梯度场,在实时VR场景下,引入加权融合的方法,得到图像的高频细节分量,利用图像融合规则,生成多尺度数字图像。实验结果表明,所提方法能够生成VR图像,并将成像真实度和清晰度分别提高到90%以上和92%以上。

面向城区的基于图去噪的小区级RSRP估计方法

移动通信系统网络的规划、部署和优化都不同程度依赖于参考信号接收功率(RSRP)估计的准确性。传统上,基站覆盖小区内某信号接收点的RSRP可由对应的无线传播模型估计。在城市环境中,不同小区的无线传播模型需要使用大量RSRP实测数据校正。由于不同小区环境存在差异,经过校正后的模型只适用于对应小区,且小区内的RSRP估计精度低。针对上述问题,将RSRP估计问题转化为图去噪问题,并通过图像处理与深度学习技术得到小区级无线传播模型,不仅能实现小区整体的RSRP估计,且能适用于相似环境小区。首先,通过随机森林回归器逐点预测每个接收点的RSRP,得到整个小区的RSRP估计图;然后,将RSRP估计图和实测RSRP分布图之间的损失视为RSRP噪声图,提出基于条件生成对抗网络(CGAN)的图去噪RSRP估计方法,通过电子环境地图反映小区的环境信息,有效地降低不同小区的RSRP。实验结果表明,在无实测数据的跨小区RSRP预测场景下,所提方法预测RSRP的均方根误差(RMSE)为6.77 dBm,相较于基于卷积神经网络的RSRP估计方法EFsNet下降2.55 dBm;在同小区RSRP预测场景下,相较于EFsNet,模型参数量减小80.3%。

针对顶点策略优化的三维模型可逆数据隐藏算法

随着3D打印技术和云计算的普及发展,三维模型的数据安全和隐私安全日益受到重视,研究模型无损恢复和存储海量数据的隐藏方法具有重要意义。基于此,以三维网格模型为载体,提出一种顶点策略优化的可逆数据隐藏方法。首先,对网格模型顶点进行划分,在全邻域中生成新的重心集合,根据改进的预测策略预测所有顶点集合的MSB以确定最大嵌入位;其次,在接受端应用顶点处拉普拉斯算子得到光滑恢复模型;最后,采用平均曲率可视化分析。为证明该算法的有效性,选取不同大小的网格模型与其他传统算法的嵌入率、SNR以及恢复模型进行对比。结果表明,该算法不仅提高了嵌入率,还在保留恢复模型局部细节特征的同时有效去除噪声,提高了模型恢复质量和视觉效果,进一步达到数据隐私安全的目的。

生成扩散模型研究综述

扩散模型在生成模型领域具有高质量的样本生成能力,一经推出就不断地刷新图像生成评价指标FID分数的记录,成为了该领域的研究热点,而此类相关综述在国内还鲜有介绍。因此,文中对相关扩散生成模型的研究进行汇总与分析。首先,对去噪扩散概率模型、基于分数的扩散生成模型和随机微分方程的扩散生成模型这3类通用模型的特点和原理进行了论述,就每一类基本扩散模型中以优化模型内部算法、高效采样为改进目标的相关衍生模型进行分析。其次,对当下扩散模型在计算机视觉、自然语言处理、时间序列、多模态和跨学科领域等方面的应用进行总结。最后,基于上述论述,分别就目前扩散生成模型存在的采样步骤多、采样时间长等局限性提出了相关建议,并结合前述研究对未来扩散生成模型的发展方向进行了研判。

基于去噪扩散概率模型的水⁃光互补系统随机场景生成方法

随着水-光互补系统应用越来越广泛,准确地描述水-光出力的不确定性对电网的运行和规划具有重要影响。现有多源融合特性建模方法不仅存在考虑可再生能源出力时空相关性不充分的问题,而且在实际复杂应用环境下要进行先验假设,进而导致生成质量降低。基于此,文中提出了基于去噪扩散概率模型的水-光互补系统随机场景生成方法。首先,将结合欧氏距离和L2正则化的损失函数作为衡量生成噪声与原始噪声分布差异的标准,并设计适应水-光-荷随机场景生成的UNet网络结构;然后,通过对前向过程不断加噪和逆向过程不断去噪训练,捕捉水-光-荷多维变量相关性变化及波动特征,拟合其概率分布规律;最后,对多源数据协同建模,高效生成水-光互补系统随机场景。文中算例基于某地区电网实际采集数据进行测试,通过综合评估指标验证了所提方法的有效性。

基于边塌陷减面的实景三维模型轻量化技术

实景三维模型由于其丰富的场景展现能力,广泛应用在数字孪生和地理测绘等领域。然而,丰富的三维场景和细节所包含的数据量十分庞大,给实景三维模型的存储、传输、渲染与展示带来了极大的挑战。针对上述问题,本文提出了一种面向实景三维模型的轻量化处理方法,首先引入模型预处理技术,去除模型噪音和非流行边;然后通过边塌陷减面的轻量化算法,精简模型拓扑结构;最后加入反转检测算法,防止边塌陷过程中出现面反转的情况。对张家界景区4个真实场景的测试结果表明,该技术可以在保留模型整体外观和场景细节的基础上,压缩至原始模型大小的3.7%~10.9%,使得复杂场景的实景三维模型能够在计算机中更加快速真实地反映或显示世界的地理地貌或城市建筑。

基于全变分正则项展开的迭代去噪网络

针对神经网络训练存在解释能力差以及不稳定问题,提出一种基于CP(Chambolle-Pock)算法求解的全变分(TV)正则项展开去噪网络(CPTV-Net),用于解决低剂量计算机断层扫描(LDCT)图像去噪问题。首先,向L1正则项模型引入TV约束项,以保留图像的结构信息;其次,采用CP算法对去噪模型进行求解并得出具体迭代步骤,保证算法的收敛性;最后,借助浅层卷积神经网络学习线性操作的原始对偶变量迭代公式,用神经网络计算模型的解,并通过收集网络参数优化合并数据。在模拟和真实LDCT数据集上的实验结果表明,与残差编码器-解码器卷积神经网络(REDCNN)、TED-Net(Transformer Encoder-decoder Dilation Network)等五种先进的去噪方法相比,CPTV-Net具有较优的峰值信噪比(PSNR)、结构相似度(SSIM)和视觉信息保真度(VIF)评估值,能生成去噪效果明显和细节保留最为完整的LDCT图像。

基于粗糙集的去噪扩散概率方法

基于非Markov链去噪扩散隐式模型(DDIM),提出一种粗糙集的去噪扩散概率方法,用粗糙集理论对采样的原序列等价划分,在原序列上构建子序列的上下近似集和粗糙度,当粗糙度最小时获取非Markov链去噪扩散隐式模型的有效子序列.利用去噪扩散概率模型(DDPM)和DDIM进行对比实验,实验结果表明,该方法获取的序列是有效子序列,且在该序列上的采样效率优于DDPM.