全变分耦合图像去噪模型

针对TV模型去噪后图像容易产生“阶梯效应”的现象,提出一种全变分耦合图像去噪模型。首先,根据去噪过程中图像梯度的变化趋势,构造一个趋势保真项,该保真项不但能有效去除图像噪声,而且能抑制“阶梯效应”。然后用小波在频域里对图像进行系数分解,利用Canny算法的边缘检测特性,设计控制函数,控制能量的扩散方向,保持了TV模型和趋势保真项的优点,能够在保护图像边缘纹理等细节信息的同时,抑制“阶梯效应”。实验结果表明,新模型的峰值信噪比、结构相似度、视觉效果均有显著提高。另外,所提模型的运行时间较短。     

基于全变分理论的红外图像去噪

为了去除红外图像中的噪声,提出了一种基于全变分理论的去噪算法。该方法继承了经典全变分模型在去除噪声中保护边缘的优点,结合图像平滑扩散原理,得到了一个全新的扩散函数;同时引入了一个边缘检测算子,对正则项和忠诚项的相关参量进行了改进,使得修复后的图像大大避免了阶梯效应;最后对该算法的实现进行了推导。结果表明,该算法能够有效地去除噪声,并且避免了阶梯效应的产生。     

基于全变分模型的图像噪声去除效果比较

基于偏微分方程的图像去噪方法由于将数学与工程结合得更加紧密,具有较强的自适应能力和灵活性.本文首先介绍了目前已经提出的变分模型的快速Split-Bregman算法,然后通过大量数值实验对不同模型的去噪效果进行了比较.所研究的模型包括L1范数、L2范数、LTV(1ayered total variation)规则项、MTV(multicharmel total variation)规则项和CTV(color total variation)规则项,从灰度图像和彩色多通道图像两方面进行分析.实验结果表明对于灰度图像基于L1范数的TV去噪模型效果较好,彩色图像中CTV模型对图像去噪边缘保持最好,其他依次是MTV模型、LTV模型.     

基于广义变分正则化的红外图像噪声抑制方法

文中提出了一种广义变分正则化的红外图像噪声抑制方法,该方法采用p-范数代替目前广泛被采用的全变分范数作为正则项,构造了用于抑制图像噪声的展平泛函,从而将图像噪声抑制问题转化为能量泛函优化问题。通过推导,得到了相应的用于图像噪声抑制的非线性偏微分方程,并采用固定点迭代算法进行线性化求解,使得迭代解稳定收敛。数值试验结果表明,该方法能够有效地去除图像噪声,较之全变分图像噪声抑制方法,新方法进一步提高了对小宽度图像边缘的保持能力,是一种有效且性能优良的红外图像噪声抑制方法。     

一种全变差正则化流场的图像抑噪方法

利用Meyer的图像分解理论,提出一种磨光流场的全变差正则化抑噪方法。该方法首先引入负指数Hilbert- Sobolev范数度量逼近项,对图像水平曲线的法向量场进行全变差正则化磨光,然后构造出一个曲面拟合模型,拟合磨光后的流场。最后,利用有限差分法对各模型所导出Euler-Lagrange方程进行数值求解。实验结果表明,该方法在有效去噪的同时,使边缘和纹理信息均得到较好的保持。     

基于深度卷积神经网络的无参考低照度图像增强

针对低照度图像增强算法在实现细节增强的同时对噪声抑制考虑的不足问题,该文提出一种基于深度卷积神经网络的无参考低照度图像增强方法。首先,基于Retinex理论从输入的低照度图像中提取照射分量和反射分量,并分别对二者进行优化,随后将优化后的照射分量和反射分量相乘得到增强后的图像;同时,将3D块匹配(BM3D)的去噪效果融合进反射分量的优化过程中;最后,采用无参考图像训练的方式,并配合改进后的趋势一致性损失对网络参数进行更新。实验结果表明,该文算法相较于现有的主流算法,可有效地提升低照度图像的对比度和亮度,同时保持图像的自然性。     

基于光谱加权低秩矩阵分解的高光谱影像去噪方法

高光谱遥感影像在获取和传输过程中会受到各种类型噪声的污染,不仅降低影像质量,也限制了其后续应用的精度。高光谱影像噪声类型复杂多样,且噪声在不同波段上的强度也并不相同。通过引入光谱域上的权重矩阵,文中提出了一种基于光谱加权低秩矩阵分解的高光谱遥感影像混合噪声去除方法,利用光谱权重矩阵均衡不同波段的噪声强度差异性。为进一步将噪声与纯净影像分离,利用加权核范数最小化来约束纯净高光谱影像的局部低秩结构,并利用交替方向乘子法对所提出的模型进行优化求解。通过对模拟与真实高光谱遥感数据的实验,验证了所提方法的有效性与优越性。     

基于结构保持的MR图像运动伪影快速抑制方法

目前核磁共振图像运动伪影的校正方法普遍是基于K空间数据的方法,本文提出一种直接对核磁共振图像进行伪影校正的后处理方法.基于非局部均值总变差去噪的思想设计构造了结构保持的运动伪影校正模型,该模型由非局部均值正则项和块相似保真项构成,正则项可以有效去除运动伪影和噪声的同时保持图像的结构;将各向异性结构张量作为块相似保真项中的权函数,实现在不同区域有不同的扩散方式,在去除图像运动伪影的同时保留图像的细节信息.模型的数值求解采用分裂Bregman方法实现.本文提出的方法充分考虑了图像的几何结构特性,实验结果表明,该方法能有效去除运动伪影并保留有价值的图像细节信息,同时提高了运算速度.     

基于L1-L2范数的正则项去噪模型的PCB图像去噪算法研究

电子行业常通过提取图像特征来对印刷电路板(Printed Circuit Board,PCB)进行缺陷识别。为了改善PCB图像的视觉效果,提升PCB无损检测的准确率,本文提出了一种基于L1-L2范数的正则项去噪模型的PCB图像去噪算法。首先采用非局部均值(Non Local Mean,NLM)滤波算法将提取的图像分解为结构和纹理两个部分,根据结构框架和纹理细节差异化的物理特性,分别使用Lasso回归算法和Ridge回归算法进行图像去噪,然后将Split Bregman迭代框架应用到去噪模型中,最后通过MATLAB软件平台对所提算法进行实验探究,并从视觉角度和去噪效果指标SNR、SSIM等多方面对算法进行评估。实验结果证明了基于L1-L2范数的正则项去噪模型的PCB图像去噪算法的有效性和可行性。     

Variational denoising of partly textured images by spatially varying constraints.

Denoising algorithms based on gradient dependent regularizers, such as nonlinear diffusion processes and total variation denoising, modify images towards piecewise constant functions. Although edge sharpness and location is well preserved, important information, encoded in image features like textures or certain details, is often compromised in the process of denoising. We propose a mechanism that better preserves fine scale features in such denoising processes. A basic pyramidal structure-texture decomposition of images is presented and analyzed. A first level of this pyramid is used to isolate the noise and the relevant texture components in order to compute spatially varying constraints based on local variance measures. A variational formulation with a spatially varying fidelity term controls the extent of denoising over image regions. Our results show visual improvement as well as an increase in the signal-to-noise ratio over scalar fidelity term processes. This type of processing can be used for a variety of tasks in partial differential equation-based image processing and computer vision, and is stable and meaningful from a mathematical viewpoint.     

A Variational Model for Staircase Reduction in Image Denoising

We propose a new variational model to reduce the staircase that often appears in Total variation (TV) based models in image denoising. The model uses BV-seminorm and Besov-seminorm to measure the piece-wise constant component and piecewise smooth component of the image, respectively. We discuss the nontrivial prop-erty of the proposed model and introduce an alternating iteration algorithm that combines the dual projection al-gorithm with Wavelet soft thresholding (WST) algorithm to solve the model numerically. The experimental results show that the proposed model is effective for noise removal and staircase reduction, while the contour can be preserved in the denoised images. Furthermore, compared with two classical staircase reduction models, CEP2 and TGV, the proposed model is much faster than these two models.     

Split Bregmanized anisotropic total variation model for image deblurring

In this paper, an effective image deblurring model is proposed to preserve sharp image edges by suppressing the stair-casing arising in the total variation (TV) based method by using the anisotropic total variation. To solve the difficult L1 norm problems, the split Bregman iteration is employed. Several synthetic degraded images are used for experiments. Comparison results are also made with total variation and nonlocal total variation based method. Experimental results show that the proposed method not only is robust to noise and different blur kernels, but also performs well on blurring images with more detailed textures, and the stair-casing effect is well suppressed.     

一种改进的非凸正则项模型的噪声去除方法

针对非凸正则项模型,在去除乘性噪声时边缘信息对噪声敏感且强度较大的噪声抑制能力弱的缺陷,提出了一种改进的图像去噪新模型。在新模型中通过取对数将乘性噪声转变成加性噪声,然后在模型的正则项和忠诚项中均引入高斯卷积,既对图像进行平滑预处理,又获得丰富的边缘信息,从而对边缘作出精确定位,使新模型具有良好的鲁棒性并根据图像的特征进行平滑,因而更好地保护了图像的边缘。数值实验表明,新方法的去噪结果在定量指标上有大幅提高,视觉效果上也有较大改善,尤其是对强度较大的噪声,新方法的优势更突出。     

Spatially adapted total variation model to remove multiplicative noise

Multiplicative noise removal based on total variation (TV) regularization has been widely researched in image science. In this paper, inspired by the spatially adapted methods for denoising Gaussian noise, we develop a variational model, which combines the TV regularizer with local constraints. It is also related to a TV model with spatially adapted regularization parameters. The automated selection of the regularization parameters is based on the local statistical characteristics of some random variable. The corresponding subproblem can be efficiently solved by the augmented Lagrangian method. Numerical examples demonstrate that the proposed algorithm is able to preserve small image details, whereas the noise in the homogeneous regions is sufficiently removed. As a consequence, our method yields better denoised results than those of the current state-of-the-art methods with respect to the signal-to-noise-ratio values.     

Fast optimization for multichannel total variation minimization with non-quadratic fidelity

Total variation (TV) has been proved very successful in image processing, and it has been combined with various non-quadratic fidelities for non-Gaussian noise removal. However, these models are hard to solve because TV is non-differentiable and nonlinear, and non-quadratic fidelity term is also nonlinear and even non-differentiable for some special cases. This prevents their widespread use in practical applications. Very recently, it was found that the augmented Lagrangian method is extremely efficient for this kind of models. However, only the single-channel case (e.g., gray images) is considered. In this paper, we propose a general computational framework based on augmented Lagrangian method for multichannel TV minimization with non-quadratic fidelity, and then show how to apply it to two special cases: L¹ and Kullback-Leibler (KL) fidelities, two common and important data terms for blurry images corrupted by impulsive noise or Poisson noise, respectively. For these typical fidelities, we show that the sub-problems either can be fast solved by FFT or have closed form solutions. The experiments demonstrate that our algorithm can fast restore high quality images.     

基于结构相似度的图像去噪新方法

将结构相似度作为一种刻画忠诚项的度量用于图像去噪模型中。针对经典ROF模型忠诚项的约束项L2度量未考虑图像空间结构性而导致恢复图像视觉效果差的缺陷,引入结构相似度来改进模型的忠诚项,提出了一种新的去噪模型。为在去噪过程中,更好地保护图像的边缘,在此模型的基础上,文中还做了进一步改进,用非凸正则项代替TV正则项,得到推广模型。实验结果表明,相对于ROF模型,两个模型在有效去除噪声的同时,能更好地保持图像的结构信息,提高图像的视觉效果,且推广模型在图像边缘保护方面的性能更好。     

医学图像增强系统设计

为了改善医学图像的视觉效果,提高图像的清晰度,使之更适合于机器的分析处理以及人的视觉特性,并突出病灶点,为病理学诊断和临床诊断提供可靠依据。设计了一个对医学图像十分具有针对性的图像增强系统。针对CT图像的电子噪声提出了基于修正维纳滤波的小波包去噪算法;针对B型超声图像的散斑噪声提出了基于脉冲耦合神经网络(PCNN)模型的小波自适应斑点噪声滤除算法;针对医学图像对比度低,边缘信息模糊等特点,提出了基于小波变换的医学图像增强算法。当噪声方差为0.01时,基于脉冲耦合神经网络(PCNN)模型的小波自适应斑点噪声滤除算法获得的PSNR比经Wiener滤波方法获得的PSNR高出9 dB。系统能快速找到噪声点进行定点去噪,能有效提高医学图像的对比度,增强边缘细节信息,突出病灶点的位置,从而达到较好的处理效果,为医疗工作者观察病症提供更加清晰准确的依据。     

基于DnCNN的海面目标一维距离像识别方法

针对低信噪比条件下海面目标分类识别精度差的问题,该文提出了一种基于去噪卷积神经网络(Denoising convolutional neural network,DnCNN)的海面目标高分辨一维距离像(High Resolution Range Profile,HRRP)识别方法。所提方法设计了一个海面目标分类识别模型,该模型通过其中的降噪模块提高信噪比。首先,分析了HRRP和二维图像的相似特性,将HRRP降噪转变为二维图像降噪。其次,利用深层次卷积层与批归一化层相结合的结构,提取图像深层次的噪声特征,最后采用残差学习技术,减轻深层次网络的学习负担的同时重构图像进行分类识别。实验结果表明,该模型可以有效提升低信噪比条件下的海面目标分类识别正确率,在不同信噪比条件下其识别性能均优于对比模型,具有良好的识别性能和鲁棒性。      

Color TV: total variation methods for restoration of vector-valuedimages

We propose a new definition of the total variation (TV) norm for vector-valued functions that can be applied to restore color and other vector-valued images. The new TV norm has the desirable properties of (1) not penalizing discontinuities (edges) in the image, (2) being rotationally invariant in the image space, and (3) reducing to the usual TV norm in the scalar case. Some numerical experiments on denoising simple color images in red-green-blue (RGB) color space are presented.     

Orthogonal-directional forward diffusion image inpainting and denoising model

In this paper, an orthogonal-directional forward diffusion Partial Differential Equation （PDE） image inpainting and denoising model which processes image based on variation problem is proposed. The novel model restores the damaged information and smoothes the noise in image simultaneously. The model is morphological invariant which processes image based on the geometrical property. The regularization item of it diffuses along and cross the isophote, and then the known image information is transported into the target region through two orthogonal directions. The cross isophote diffusion part is the TV （Total Variation） equation and the along isophote diffusion part is the inviscid Helmholtz vorticity equation. The equivalence between the Helmholtz equation and the inpainting PDEs is proved. The model with the fidelity item which is used in the whole image domain denoises while preserving edges. So the novel model could inpaint and denoise simultaneously. Both theoretical analysis and experiments have verified the validity of the novel model proposed in this paper.