Implementation decision making for internetware driven by quality requirements

Internetware is an emerging software paradigm in the open,dynamic and ever-changing Internet environment.A successful internetware must demonstrate acceptable degree of quality when carrying out its functionality.Hence,when internetware is being dynamically constructed,making implementation decisions to satisfice the quality requirements becomes a critical issue.In the traditional software engineering,quality requirements are usually refined stepwise by sub-requirements utilizing goal modeling perspective,until some potential functional design alternatives are identified.The goal-oriented paradigms have adopted graphical goal models to reason about quality requirements and proposed qualitative or quantitative reasoning schemas.However,these techniques may become unviable due to the ever-changing operating environment and demands for run-time decision making.In this paper,we propose an approach for implementation decision making driven by quality requirements for internetware.It focuses on the symbolic formula representation of requirements goal models with the tree structure,which is of well-defined syntax and clear traceability.Furthermore,we explore some reasoning rules which effectively automate each reasoning action on the formulae.This supports multiple-factor decision making.A case study is also provided to illustrate our proposed approach.We have developed a supporting tool based on our theoretical approach that we also present in this paper.     

A novel approach for multi-scale texture segmentation based on fractional differential

In this paper, we intend to implement multi-scale texture segmentation by fractional differential. We propose two fractional differential masks and present the structures and parameters of each mask, respectively, on eight directions. Moreover, by theoretical and experimental analysis, we find the better performance fractional differential mask. Finally, we further discuss the capability of fractional differential for multi-scale texture segmentation. Experiments show that, for rich-grained digital images, the capability for multi-scale texture segmentation by fractional differential-based approach appears efficient.     

A texture image denoising approach based on fractional developmental mathematics

The traditional integer-order computation-based denoising approaches often blur the edges and textural details of an image. To solve this problem, from the viewpoint of system evolution, and based on the features of fractional calculus, we propose to implement a texture image denoising approach based on fractional developmental mathematics (FDM) which applies a novel mathematical method, fractional calculus, to image denoising. First, we synopsize the necessary theoretical background of fractional calculus. Second, we derive the necessary mathematical models for implementation of a texture image denoising approach based on FDM. We derive fractional Green’s formula, fractional Gauss’ formula, and fractional Stokes’ formula, and fractional Euler–Lagrange equation. Then, a texture image denoising approach based on FDM is proposed. Third, we implement comparative experiments. We firstly derive the numerical implementation of FDM. Then, we study the capability of preserving the edges and textural details of FDM by comparative experiments. The comparative experimental results show that the capability of preserving the edges and textural details of the FDM-based denoising algorithm is obviously superior to that of traditional integer-order computation-based algorithms, especially for texture detail rich images.     

基于分数阶微分的超声斑点去噪

为了保留更多的纹理信息,构建了基于具有阻止扩散的梯度阈值k,和分数阶微分的阶数v平衡关系的分数偏微分方程的图像去噪模型,其有效结合了分数微积分理论和偏微分方程方法,并通过分数微分掩模算子实现了数值.超声体模信验和体内成像表明:基于分数阶微分的各向异性扩散方法可以提高组织的信噪比(SNR)和超声图像的质量.     

基于分数阶偏微分方程的图像去噪新模型

将分数阶微分理论和全变分方法相结合应用于图像去噪,提出了一种基于分数阶偏微分方程的图像去噪新模型。该模型很好地继承了现有的全变分(TV)模型去噪效果与保持图像边缘细节特征的优点,同时利用分数阶微分运算特有的幅频特性优势,较好地保留了图像平滑区域中灰度变化不大的纹理细节。实验结果表明：一方面,与现有去噪方法相比,新模型不仅具有较强的抑制噪声能力,而且能较好地保持图像边缘特征,还能保留更多的图像纹理细节信息,优于常用的整数阶偏微分图像去噪方法;另一方面,从峰值信噪比的对比实验可以看出该模型去噪效果优于其他方法,较好地达到了去噪目的,是一种有效、实用的图像去噪模型。     

分数阶B样条小波域的图像变分去噪

分数阶B样条具有分数阶逼近,可以更好地刻画图像纹理部分。将分数阶B样条小波推广到二维领域,利用分数阶B样条小波进行图像阈值去噪,提出了分数阶B样条小波域图像去噪的变分模型。同传统小波函数与全变差结合模型比较,分数阶B样条小波在保持纹理和去噪方面得到了明显改进。     

分数阶偏微分方程在图像处理中的应用

分数阶偏微分方程在图像处理中的应用已受到了广泛的关注,尤其在图像去噪和图像超分辨率（SR）重建方面,目前的研究成果已显示了分数阶应用的优势与效果。对分数阶微积分在图像处理中的作用进行了分析;介绍并讨论了分数阶偏微分方程在图像去噪和图像超分辨率重建中的相关理论与模型;通过仿真实验表明,基于分数阶偏微分方程的方法在去噪和减少阶梯效应等方面比整数阶偏微分方程更具有优势;最后指出了未来的相关研究问题。     

分数阶微积分在图像处理中的研究综述*

综述了关于分数阶微积分理论在数字图像底层处理中的应用研究,具体包括:分数阶微积分和分数阶偏微分方程的基本理论及分数阶傅里叶变换的基本性质;分数阶微分滤波器的构造及在图像增强中的应用研究;分数阶积分滤波器的构造及在图像去噪中的应用研究;分数阶偏微分方程在图像去噪中的应用研究。最后,总结了分数阶微积分理论在图像处理中已取得的研究成果,并结合已有的基于分数阶微积分理论的图像底层处理模型,展望了分数阶微积分理论在图像处理中的应用前景。     

基于分数阶Riemann-Liouville积分的图像去噪

为了在获得更好去噪性能的同时更多地保留图像纹理信息,介绍了分数阶Riemann-Liouville(R-L)积分算子在信号滤波中的作用,将分数阶R-L积分理论引入到数字图像去噪中,并利用阶梯逼近方法来实现数值计算。模型通过设定微小的积分阶次来构建相应的图像去噪掩模,由此实现噪声图像的局部微调,并利用迭代的思想来控制模型的去噪强度,从而获得较好的图像去噪效果。实验结果表明,基于分数阶R-L积分的图像去噪算法较传统的去噪方法不仅可以提高图像的信噪比(SNR),所提出的算法去噪后图像的信噪比为18.3497dB,较传统去噪方法最低也提升了大约4%,而且可以更好地保留图像的弱边缘和纹理等细节信息。     

基于分数阶微分自适应算法的煤尘图像滤噪

针对传统的煤尘图像滤噪方法迭代过程长、滤噪效果不理想、纹理保持能力差等问题,对现有的滤噪方法进行改进,建立了基于分数阶微分模型的自适应滤噪算法。改进算法对参数u的变化梯度进行调整,从整数阶扩展到分数阶;根据区域特征分别对算法中的各项参数进行自适应选择。实验结果表明,改进后的滤噪算法收敛速度快,迭代次数少,滤噪效果好,纹理保持能力强,且其检测滤噪效果能力的量化指标获得了很好的改善。     

基于分数阶变分的图像泊松去噪模型

为了进一步提高图像去噪的效果,针对图像泊松噪声的特点,提出了一种有效的基于分数阶导数的图像泊松去噪的变分模型。该模型继承了全变分模型去噪效果良好的优点,并且很好地利用分数阶微分特有的幅频特性优势,在处理图像细节和纹理特征方面很好的保留了图像的“弱信息”。数值实验结果表明,该分数阶变分方法的去噪效果优于传统的整数阶变分方法,能很好地保留图像的边缘细节特征。     

基于有理数阶微分的图像去噪新方法

针对现有的全变分(TV)去噪方法效果不太理想,在去噪的同时不能较好地保持图像的边缘和纹理细节,提出了一种基于有理数阶微分的图像去噪新方法。首先详细地讨论了现有的全变分去噪方法和分数阶微分去噪方法各自的优缺点;然后将全变分去噪模型与分数阶微分理论相结合,获得有理数阶微分图像去噪新模型,并推导了相应的有理数阶微分模板。实验结果表明：与改进前的方法相比,信噪比(SNR)提高了接近2个百分点,较好地传承了全变分去噪方法对图像高频部分大幅改善及分数阶微分去噪方法能够很好地保留图像纹理细节的优点,是一种有效的图像去噪方法。     

Riemann Liouville分数阶积分的图像去噪算法

传统图像去噪算法易丢失图像边缘和纹理细节,使图像模糊不清,为后续图像分析处理带来困难。为克服传统图像去噪算法的缺点,根据Riemann Liouville分数阶积分,构造一种分数阶积分掩膜算子,对测试图像进行图像去噪仿真实验。同时,引入客观评价标准峰值信噪比和灰度共生矩阵,对分数阶积分掩膜算子的去噪效果进行分析。结果表明,不同于传统图像去噪算法,该分数阶积分掩膜算子可在去除图像噪声的同时,有效保留图像的边缘和纹理细节信息。     

具有纹理保持能力的四阶偏微分方程去噪方法

虽然四阶偏微分方程图像去噪方法能得到较好的分段光滑的结果,但这类方法常破坏图像的纹理信息。提出了一种具有保持图像纹理信息能力的四阶偏微分方程去噪模型。利用垂直于梯度方向的图像二阶导数设计了一种新的代价函数。证明了该函数解的存在性与唯一性并给出了其对应的Euler-Lagrange方程。在实验方面,用大量真实的纹理图像验证了新方法。实验结果表明,新方法在去噪的同时图像的边缘与细节得到了较好的保持。     

一种自适应分数阶偏微分图像增强模型

针对传统的自适应分数阶偏微分方程图像增强算法对图像暗区纹理区域的增强不足的缺点,考虑到人眼对光感的敏感程度不同,将亮度对视觉的影响因素考虑进传统的自适应分数阶偏微分方程图像增强算法。以梯度和灰度值为参数,建立了一种新的自适应分数阶偏微分图像增强模型。该模型改善了传统算法对暗区图像增强不足的缺点,图像增强后的平均梯度提升明显,很好地改善了图像的视觉效果。实验结果说明本算法具有一定的有效性。     

基于分数阶偏微分方程的彩色图像去噪新方法

为了克服现有彩色图像去噪方法不能有效抑制边缘噪声和保持纹理信息的缺点, 提出了一种结合了离散四元数傅里叶变换（DQFT）和分数阶微分理论的彩色图像去噪改进方法。算法采用四元数矩阵表示一幅彩色图像, 首先对该四元数矩阵进行离散傅里叶变换; 然后将其代入基于分数阶微分的能量泛函极小值求解过程中, 利用变分原理求解并推导出去噪模型, 并与两种传统彩色图像去噪模型进行实验比较。实验结果表明, 提出的模型在去噪效果和纹理保护方面都有更好的表现。     

基于改进阈值函数的分数阶小波图像去噪

为了消除噪声对图像的影响并较好地保留图像细节信息,提出一种基于改进阈值函数的分数阶小波图像去噪方法。该方法通过分数阶小波变换将含噪信号进行多尺度分解,采用改进的阈值函数对各层分数阶小波域系数进行处理,对处理后的系数进行重构得到去噪后的信号。仿真实验表明,相比已有的软阈值、硬阈值和均值加权法,本文方法去噪后的图像信噪比较大、均方误差较小,取得了满意的视觉效果,是一种实用的去噪方法。     

自适应残差图像的分数阶全变分去噪算法

目的 全变分（TV）去噪模型具有较好的去噪效果,但对于图像的弱边缘和纹理细节的保持不够理想。自适应分数阶全变分（AFTV）模型根据图像局部信息,区分图像的纹理区域和非纹理区域,自适应计算投影算法中的软阈值,可较好地保持图像的弱边缘和纹理细节,但该方法当噪声增大时“阶梯”效应比较明显,弱边缘和纹理细节保持效果不够理想。针对该问题,提出一种改进的分数阶全变分去噪算法。方法 该算法在计算残差图像时,用分数阶全变分模型替代整数一阶全变分模型,并根据较精确的残差图像的局部方差区分图像纹理区域和平坦区域,使保真项参数的自适应选取更加合理,提高了算法的去噪性能。结果 针对3种不同类型的噪声图像,将本文模型与TV模型和AFTV模型进行对比实验,并采用峰值信噪比（PSNR）和结构相似性（SSIM）评定去噪效果和纹理保持能力。对于高斯噪声图像,本文算法在PSNR方面比TV模型和AFTV模型分别可平均提高2.72 dB和1.38 dB,SSIM分别可平均提高0.047和0.020。对于椒盐噪声图像,本文算法结合中值滤波算法在PSNR和SSIM方面比传统中值滤波算法分别可平均提高1.308 dB和0.011。对于泊松噪声图像,本文算法在PSNR、SSIM方面与AFTV较接近,比TV分别可提高1.59 dB和0.005。结论 通过对添加不同类型的噪声图像进行实验,结果表明提出的算法在去噪性能上与TV和AFTV相比均有较大提高,尤其对于噪声较大的图像效果更为显著,在去噪效率上与AFTV的时间复杂度相当,时耗接近略有降低。且本文算法普适性较好,能有效去除多种典型类型的噪声。     

Using reproducing kernel for solving a class of time-fractional telegraph equation with initial value conditions

Today, most of the real physical world problems can be best modelled with fractional telegraph equation. Besides modelling, the solution techniques and their reliability are the most important. Therefore, high accuracy solutions are always needed. As we all know, reproducing kernel method (RKM) has been successfully presented for solving various ordinary differential equations. However, the numerical results are not perfectly satisfactory when we directly use the traditional RKM for solving fractional partial differential equation. The aim of this paper is to fill this gap. In this paper, a new method is provided for solving fractional telegraph equation in the reproducing kernel space by piecewise technique, which can obtain more accurate solution than traditional method. Three experiments are given to demonstrate the effectiveness of the present method.     

On parameter identification for distributed systems using Galerkin's criterion

A method is presented for estimating parameters in distributed parameter systems. The system is assumed to be modeled by a set of partial differential equations whose form is known to within a set of unknown constant parameters. Galerkin's Method is used to transform the partial differential equations into a set of ordinary differential equations. The approach to the identification problem is given in a step by step procedure. Three optimization schemes for estimating the unknown parameters are discussed. They are a steepest descent method, a search technique, and nonlinear filtering.