基于纹理特征的三维点云数据配准算法研究

使用Kinect可以方便地获取物体的纹理图像和三维点云数据。研究一种通过获取纹理图像的特征点进行快速三维点云数据配准的算法．并最终应用到室内环境的三维场芾重建中。实验表明,此算法具有直观、实现简单、运算量小等优点。     

基于纹理分类的极化SAR图像滤波方法

关于雷达图像优化,提高分辨率的问题,场景较为复杂的图像,固有噪声图像效果不够理想,对具有不同统计特性的像素点缺乏精确的区分。由于传统参数估计方法降噪效果不足,为解决上述问题,提出了一种基于纹理特征分类的参数估计方法。首先计算极化总功率图像的灰度共生矩阵,并提取纹理特征矢量,用K均值聚类的方法进行分类。然后根据分类结果,在滑动邻域窗内选取与中心像素同类别的像素用于参数估计。实验结果表明,改进的纹理分类的滤波方法具有更好的降噪效果,对于复杂场景的极化SAR图像表现了较大的优越性。     

一种改进的人脸纹理映射方法

识别人脸图像优化问题,由于人脸纹理映射时容易出现的纹理扭曲和畸变现象,影响图像的真实性,提出一种改进的纹理图像映射方法。根据人脸特征点信息,利用特征块匹配法配准正侧面人脸图像,并使用加权平滑算法和金字塔方法对图像进行双重融合实现图像的平滑过渡;然后利用一种特殊的定位方法来计算三维人脸模型上顶点对应的纹理坐标;最后根据纹理坐标将纹理图像映射到三维特定人脸模型表面上,可得到具有真实感的三维人脸模型。实验结果表明,纹理合成方法简捷有效,纹理映射结果生动逼真,有效的避免了纹理图像的扭曲和畸变现象,具有较高的鲁棒性。     

PVA-based activated carbon fibers with lotus root-like axially porous structure

A novel method was developed for the fabrication of activated carbon fiber (ACF) with wet spinning polyvinyl alcohol (PVA) fibers as the precursors. Through a combination of preoxidation, dehydration, carbonization and activation under a certain tension, PVA-based ACFs (PVA-ACFs) with high yields and good mechanical properties were obtained. The surface and cross-section morphologies, pore structures, surface geometries, surface functional groups and crystal structures of the PVA-ACFs were characterized using field emission scanning electron microscopy, low temperature nitrogen adsorption, Fourier transformed infrared spectrophotometry, X-ray photoelectron spectroscopy and X-ray power diffraction. All the PVA-ACF samples prepared had lotus root-like axially meso- and macroporous structures with the domination of micropores. Fractal geometries of PVA-ACFs deduced from nitrogen adsorption isotherms indicates that capillary force dominated in the interactions between nitrogen and the PVA-ACFs. The surface functional groups of the PVA-ACFs depend on the activating agents. Graphitoidal crystal structures were observed for the PVA-ACFs. The small crystal size and short range ordering between the crystallites ensured a high specific surface area of the PVA-ACFs.     

Carbon nanotubes prepared from three-layered copolymer microspheres of acrylonitrile and methylmethacrylate

Carbon nanotubes (CNTs) were synthesized from fine three-layered copolymer microspheres using the polymer blend technique. Diameter of PMMA core/Poly(AN-co-MMA) shell-1/PMMA shell-2 microspheres, prepared by a radical soap-free emulsion polymerization of methylmethacrylate (MMA) and acrylonitrile (AN), was between 400 nm and 500 nm. Microspheres were subjected to melt-spinning at 305 °C, stabilizing in oxygen at 220 °C for 4 h, and finally carbonizing at 1000 °C for 30 min. FE-SEM study of carbonized sample revealed the presence of CNTs arrays on carbon blocks. Similar arrays were observed in a comparative CNTs sample prepared from three-layered microspheres with the pure PAN shells-1 layers. HRTEM showed that the CNTs derived from copolymer microspheres had different structure when compared to the control sample, i.e. CNTs often adhered to each other and contained the internal compartments. The insufficient PMMA shell-2 coating of copolymer microspheres is believed to be a reason for CNTs adhesion. The possible mechanisms of the carbon block formation and the adhesion of CNTs are introduced.     

基于局部重要性采样的SAR图像纹理特征提取方法

合成孔径雷达（Synthetic aperture radar,SAR）图像因为相干斑现象和目标响应的空间变化呈现出一种纹理特性,局部二进编码等局部图像特征 在光学纹理描述中获得较好的结果,但光学纹理特征在描述SAR图像纹理特性中因为相干成像特性往往失效. 本文 在前期工作纹理特征框架的基础上,提出了一种局部重要性采样二进编码的SAR图像纹理特征（Feature extraction based on local important sampling binary,LISBF）描述方法:首先,利用样本图像对局部采样位 置进行随机自适应采样,基于重要性采样（Important sample,IS）方法输出递归学习位置结果;然后,利用学习出的纹理重要采样点对进 行二进特征编码;最后,通过映射和统计生成描述算子. 该特征较固定位置采样能够获取更大范围信息,同时能通 过采样避免特征维数的急剧增大;通过自适应学习重要性关键点较随机采样 更容易捕捉纹理固有信息;较好地适应了SAR图像极低信噪比和斑点现象的纹理. 本文将该特征用于真实图像和标准纹理库的分类研究,实验结果证明了该特征的有效性.     

纹理豪斯多夫距离在人脸识别算法中的应用

很多经典的人脸识别方法难以适应姿势变化及人脸错位的情形。为了解决这一问题,提出一种基于纹理豪斯多夫距离(THD)的人脸识别算法。将人脸图像的空间量及纹理特征相结合,使其在深入的头部转动和人脸错位中都有很高的容错度。在FERET及Yale两大人脸数据库的实验表明,与其他经典的方法相比较,所提出的方法取得了更好的识别效果。     

雕塑创作中材质肌理的运用

主要分析了材质肌理在雕塑创作中的运用，通过我对泥、木、石、金属及其肌理表现的深层次的认识与分析，结合许多实例，来说明材质肌理在雕塑创作中的重要性。     

添加石英对莫来石－刚玉制品结构的影响

探讨了引入石英颗粒的莫来石－刚玉制品中大气孔的成因。认为与液相的生成及其在毛细管力作用下的转移有关；与莫来石生成膨胀效应对颗粒拉紧的阻碍有关；与石英相交所引起的体积膨胀有关。     

Constructing and applying higher order textons: Estimating breast cancer risk

Texture analysis based on textons is extended by introducing a method for computing textons of arbitrary order. First-, second- and third-order textons are applied to classify screening mammograms as to indicate a low or high risk of breast cancer. First-order textons are found to provide better estimates of breast cancer risk than other orders on their own but the combination of first- and second-order textons outperforms first-order textons alone and other combinations of two orders. Combining all three orders of textons does not improve classification. This example indicates that including higher-order textons has the potential to improve classification performance.