基于变换域VGGNet19的红外与可见光图像融合

针对红外与可见光图像融合中出现细节信息丢失及边缘模糊的问题,提出一种在变换域中通过VGGNet19网络的红外与可见光图像融合方法。首先,为了使得源图像在分解过程中提取到精度更高的基础与细节信息,将源图像利用具有保边平滑功能的多尺度引导滤波器进行分解,分解为一个基础层与多个细节层;然后,采用具有保留主要能量信息特点的拉普拉斯能量对基础层进行融合得到基础融合图;其次,为了防止融合结果丢失一些细节边缘信息,采用VGGNet19网络对细节层进行特征提取,L1正则化、上采样以及最终的加权平均策略得到融合后的细节部分;最后,通过两种融合图的相加即可得最终的融合结果。实验结果表明,本文方法更好地提取了源图像中的边缘及细节信息,在主观评价以及客观评价指标中均取得了更好的效果。     

基于改进滤波和标记提取的分水岭算法

针对彩色图像分割中分水岭算法的过分割问题,提出了一种改进的基于标记提取的分水岭算法.改进后的算法由平滑滤波、彩色梯度计算、标记提取和分水岭变换组成.在平滑滤波阶段,设计了保边性能优于传统频域低通滤波器的频谱包络滤波器并运用于彩色图像及其梯度的平滑.彩色图像梯度计算直接在彩色向量空间进行.在标记提取阶段,利用局部极小值区...     

结合图像分割的改进导引滤波

为了更好地对图像进行平滑同时保持边缘不模糊,设计一种新的滤波方法。对基于该方法的图像滤波、细节增强等算法进行研究。首先,根据图像的亮度和颜色对图像进行分割,将图像分成不同的区域。接着,在不同的区域进行导引滤波,得到互不交叠的多个滤波子图像。然后,将这些子图融合,得到基于分割的改进导引滤波结果。最后,利用提出的改进导引滤波方法提出一种多尺度的细节增强方法。实验结果表明,在图像光滑和细节增强方面,提出的方法都要好于传统的导引滤波:提出的方法不仅能较好的光滑图像,同时保持边缘清晰,减少了传统滤波方法在边缘处的光晕现象。     

In the name of art

In building a digital library of Chinese calligraphy, it is desirable to preserve the dynamic writing as well as the aesthetic properties of the masterpieces. Compared with the work on computer synthesis of Chinese calligraphy, we are dealing with the reverse problem; it can be achieved by performing image analysis on the archaic Chinese calligraphic images to obtain a model-based parametric representation. In this article, we introduce our methodology to preserve archaic Chinese calligraphic images by automatically estimating the set of three-dimensional (3-D) geometric and ink model parameters such as the properties of the brush hair and the variations of ink deposition along the stroke trajectory. The dynamic brush writing of the archaic Chinese characters can then be reproduced with various brush and ink effects so that the calligraphy generation process can be visualized and the masterpieces can be reproduced with high-quality printing in scalable resolution.     

基于遗传算法与单纯形下山法的广义切比雪夫滤波器的优化设计

广义切比雪夫滤波器是一种高性能滤波器形式,用经典的综合方法提取其参数复杂且难以实现。本文以指定的电路拓扑结构和传输零点作为已知条件,结合遗传算法和单纯形下山法用MATLAB编程实现了一种新的灵活、有效的广义切比雪夫滤波器等效电路参数的优化提取方法。并给出了归一化耦合系数的物理意义及其工程应用系数转换公式。     

基于Wigner分布的脑电信号处理

临床实践表明,脑电信号中包含有大量的生理与疾病信息。对脑电信号进行行之有效的处理,不仅可以为医生提供临床诊断信息,而且可以为某些脑疾病的治疗提供有效的治疗手段。目前,随着信号处理技术的发展,在脑电信号处理中已应用了多种信号分析方法来提取脑电信号中所包含的信息,但大多数还是停留在理论研究阶段。本文在研制虚拟式脑电图仪的过程中,考虑到Wigner分布在各种时频分布中具有最简单的形式和良好的性质,从临床应用及医学研究相结合的角度出发,应用Wigner分布对脑电信号进行时频分析以提取脑电信号中的特征信息。对实测脑电数据的分析表明,应用此方法可获得较好的分析效果。     

Design of Sigma-Point Kalman Filter with Recursive Updated Measurement

In this study, the authors focus on improving measurement update of existing nonlinear Kalman approximation filter and propose a new sigma-point Kalman filter with recursive measurement update. Statistical linearization technique based on sigma transformation is utilized in the proposed filter to linearize the nonlinear measurement function, and linear measurement update is applied gradually and repeatedly based on the statistically linearized measurement equation. The total measurement update of the proposed filter is nonlinear, and the proposed filter can extract state information from nonlinear measurement better than existing nonlinear filters. Simulation results show that the proposed method has higher estimation accuracy than existing methods.     

基于双引导滤波的红外和可见光图像融合算法

图像融合是将多幅图像中有用或互补信息整合成一幅图像的过程。本文提出了一种基于引导滤波多尺度分解的红外和可见光图像融合算法。在传统的引导滤波图像融合算法的基础之上,利用双引导滤波器代替均值滤波器将源图像分解为小尺度纹理细节、大尺度边缘和基础图像;直接利用纹理细节及边缘层图像构建显著性映射图,用其代替额外的特征提取操作,可很好地突出源图像显著性信息的同时大大降低算法复杂度;利用显著性映射图、Sigmoid函数构造权重图,将源图像中具有视觉意义的信息注入到融合图像中;利用色彩模型转换融合方式,可更好保留图像的色彩信息。定性和定量实验结果证明,相比于传统的基于引导滤波的图像融合算法,本文算法的融合效果得到进一步提升。     

基于FGF与VGGNet19的红外可见光图像融合

针对红外与可见光图像融合中出现红外目标不够突出以及边缘模糊等问题,本文提出了一种基于FGF(快速引导滤波)和VGGNet19的红外与可见光图像融合方法。首先,通过FGF将源图像进行双尺度分解；然后,利用VGGNet19网络得到初步融合权重图；其次,利用OTSU(最大类间方差法)来提取初步融合权重图的前景与背景,进行二值化；然后,对得到的二值图像使用FGF得到最终融合权重图；最后,相加再经过FGF和图像增强实现图像融合。实验结果表明,本文方法能够有效保留显著热目标,边缘细节清晰,在主观评价以及客观评价指标中均取得了更好的效果。     

传统元素在Flash实验动画创作中的应用研究

Flash是美国的MACROMEDIA公司于1999年6月推出的优秀多媒体动画设计软件。随着网络平台的增多,网络信息科技水平提升,影片动画产业日益壮大。在Flash软件制作动画这一领域中,中国动画业的发展受到国外动画业发展所压迫,要想发展就必须要走出自己的路。中国的动画影片渗透“民族文化”,中国的传统元素融合在短片创作中形成一种独特的艺术风格,形成专属民族内涵的视觉文化,探索将flash动画和传统文化里的不同元素结合,形成一种新的方向,创造出富有时代风貌和民族特色的艺术动画模式。     

Unfeelable Mechanical Cloak Based on Proportional Parameter Transform in Bimode Structures

A bimode structure is a kind of structure with a high ratio of bulk moduli to shear moduli, which has a Poisson's ratio near 0.5 and a simple‐form elastic tensor. For structures in case of defects, there will be stress concentration and even failure. In this work, a method by proportional transformation for geometry parameters of a hexagonal bimode structure around the defect hole based on the special property of this kind of structure and invariant effective stiffness of a double‐trapezoid is proposed. This method can extremely reduce stress concentration caused by defects in the case of the same size of hole in this structure, making the hole not detectable by external forces. For a typical structure with a circular hole, the strain can be reduced from 378.5% to 114.8% relative to the nonhole case (and can be more effective in other parameters), and this method can also be used in other‐shape‐hole cases. This transformation method provides a direct platform for designing the nonstress‐concentration structure under defect holes, and besides making the structure lightweight.     

基于FRGF和改进PCNN的红外可见光图像融合

为解决图像融合中边缘细节保留不理想的问题,本文提出了一种快速滚动引导滤波器和改进脉冲耦合神经网络相结合的红外可见光图像融合方法。提出的快速滚动引导滤波器可以较好地在保留边缘、细节纹理信息的同时有效提高运行效率。首先,利用快速滚动引导滤波和高斯滤波对源图像进行多尺度分解;其次,为了使基础层图像更好地突出轮廓信息,采用相似性匹配的融合规则对图像进行融合;然后,细节层采用改进参数自适应脉冲耦合神经网络规则进行融合;最后,经过多尺度重构得到融合结果图。实验结果表明,与其它5种融合方法相比,该算法不仅在视觉效果上得到了提升,而且能够充分保存图像的边缘和纹理等信息,极大地提高了运行效率。另外,该方法在客观评价指标上均优于对比方法。     

基于广义B样条滤波器的图像边缘检测

为了更准确有效地提取图像边缘,提出一种基于广义B样条数字滤波器的边缘检测算法.首先由线性微分方程推导出广义B样条的一般形式.其后,利用广义B样条函数组建了边缘检测微分算子模板,该模板继承了广义B样条尺度因子α,通过调节α,可以改变边缘算子滤波器的幅值特性和带通特性,进而获得边缘表征的最佳效果.而后,结合变分公式和广义B样条函数构造广义B样条光滑滤波器,该滤波器实现了在光滑滤波意义下的直接样条变换,与边缘检测模板算子配合使用可提高边缘检测的抗噪能力.实验证明：该边缘检测算法能有效地检测出图像边缘,无论是边缘提取效果还是抗干扰能力都优于传统的差分算子.     

Pade table, continued fraction expansion, and perfectreconstruction filter banks

We investigate the relationships among the Pade table, continued fraction expansions and perfect reconstruction (PR) filter banks. We show how the Pade table can be utilized to develop a new lattice structure for general two-channel bi-orthogonal perfect reconstruction (PR) filter banks. This is achieved through characterization of all two-channel biorthogonal PR filter banks. The parameterization found using this method is unique for each filter bank. Similar to any other lattice structure, the PR property is achieved structurally and the quantization of the parameters of the lattice does not effect this property. Furthermore, we demonstrate that for a given filter, the set of all complementary filters can be uniquely specified by two parameters, namely, the end-to-end delay of the system and a scalar quantity. Finally, we investigate the convergence of the successive filters found through the proposed lattice structure and develop a sufficient condition for this convergence     

基于信息量制约的遥感图像融合算法

为了改善遥感图像的融合质量,制定了信息量制约法则用于获取融合遥感图像。利用亮度－色调－饱和度(IHS)变换从多光谱(MS)图像中获取亮度(I)分量。基于非下采样剪切波变换(NSST),获取I分量与全色(PAN)图像完成分解,得到I分量与PAN图像的高低频系数。最后,利用低频系数的均值以及区域能量信息,建立信息量制约法则,分析不同低频系数对融合系数的影响程度,设计不同的融合方法对这些低频系数进行融合。采用改进的平均梯度度量模型完成高频系数融合。通过NSST逆变换对计算融合后的高低频系数,输出新的亮度分量,将其与原始的色调(H)、饱和度(S)分量组合,通过IHS逆变换,获取融合结果。实验数据显示,相对已有的融合算法,所提算法的融合图像所含的信息量更为丰富,且光谱扭曲度更小。     

双色红外信号亚成像目标识别处理系统研究

采用高速数字信号处理器(DSP)软硬件的技术途径，在双色红外制导导引头信号处理中，对由红外传感器不同波段提供的大量目标信息，进行了综合分析并提取目标特征量。应用目标识别算法，灰度直方图变换、多尺度变换、图像滤波等方法区分真假目标，建立判决理论，确立逻辑选择条件。在算法编程，实现对目标的亚成像图及目标方位信息的准确判定等方面，作了一定的研究与探讨。     

基于总体变差模型的数字滤波器设计及其性能研究

根据经典总体变差恢复模型,设计了一种新型的应用于图像去噪的数字TV滤波器的一种算法。该新型滤波器是绝对稳定的低通滤波器,在抑制噪声的同时能够很好的保留图像的细节信息。与一般的统计滤波器不同,该滤波器充分利用了局部数据的梯度信息精确计算滤波器系数,滤波过程是图像自适应的。文中直接给出了新型滤波器固定的模板形式,使滤波过程简单易行。最后将新型滤波器与中值滤波器、均值滤波器、高斯滤波器等进行性能比较,效果更好。     

A Vector-Fitting Formulation for Parameter Extraction of Lossy Microwave Filters

In this letter, a formulation based on the vector fitting is applied to extract the equivalent circuit model from the frequency response of lossy cross-coupled microwave filters. By approximating the lossy response with short-circuit admittance parameters in partial fractional expansion form, the proposed method can evaluate the unloaded quality factor of resonators and extract the transversal coupling matrix simultaneously. The methodology of the vector fitting can identify the poles and residues of the short-circuit admittance parameters even when the poles are on the complex plane. And the extracted transversal coupling matrix can further transform into the prescribed form corresponding to the physical layout. The proposed method can be used in the tuning process of filter designs where the extraction of a coupling matrix is essential. To verify the method, a cross-coupled quadruplet filter is used as an example     

Latent Style Model: Discovering writing styles for calligraphy works

Chinese calligraphy works is a valuable part of the Chinese culture heritage. More and more calligraphy works images are digitized, preserved and exhibited in digital library. Users always want to appreciate the style-similar works simultaneously. To satisfy their need, calligraphic style representation and browsing calligraphy works by its style are the most important problems to be addressed. This paper proposes calligraphic style representation which is a multinomial probability distribution over visual words, and Latent Style Model to discover the style of calligraphy works and organize the works by style. In our experiments, we evaluated various factors that influence the model, and proved the effectiveness of the style representation and the model. At last, we illustrate the Calligraphic Style Browser to organize and exhibit the resource according to the styles.     

基于引导滤波的高动态红外图像压缩算法

非制冷红外机芯的原始图像位宽为14 Bit,虽然相对于8 Bit红外图像来说,14 Bit红外原始图像具有灰度动态范围更广、灵敏度更高、包含细节信息更多的优点,但是由于普通显示设备能够显示的最大灰度范围为8 Bit,所以需要对14 Bit原始图像进行压缩,以满足常规显示设备显示及后端图像处理的需求。若压缩算法性能不佳,在图像压缩过程中可能会丢失大量的细节信息,直接影响成像质量。本文提出了一种基于引导滤波的压缩和显示算法,该算法首先利用引导滤波对图像进行分层,根据不同图层的特点和在合成图像中的作用,分别进行增强和降噪处理,使合成图片具有良好的显示效果。通过与常用算法的仿真对比实验,以及对视觉效果和定量评价参数两个方面的实验结果分析,本文所提出算法在压缩和图像处理的性能表现上均获得一定程度的提升。