新型复合有机膨胀剂的应用

通过对腐殖酸和木素磺酸盐进行复合试验,确定了其复合比例为1∶1～3。经过蓄电池测试验证,这一复合比例提高了蓄电池的-18℃冷起动性能。     

Robust pose invariant face recognition using coupled latent space discriminant analysis

We propose a novel pose-invariant face recognition approach which we call Discriminant Multiple Coupled Latent Subspace framework. It finds the sets of projection directions for different poses such that the projected images of the same subject in different poses are maximally correlated in the latent space. Discriminant analysis with artificially simulated pose errors in the latent space makes it robust to small pose errors caused due to a subject’s incorrect pose estimation. We do a comparative analysis of three popular latent space learning approaches: Partial Least Squares (PLSs), Bilinear Model (BLM) and Canonical Correlational Analysis (CCA) in the proposed coupled latent subspace framework. We experimentally demonstrate that using more than two poses simultaneously with CCA results in better performance. We report state-of-the-art results for pose-invariant face recognition on CMU PIE and FERET and comparable results on MultiPIE when using only four fiducial points for alignment and intensity features.     

基于CCA的人耳和侧面人脸特征融合的身份识别*

鉴于人耳和人脸特殊的生理位置关系,从非打扰识别的角度出发,提出仅采集侧面人脸图像,利用典型相关分析的思想提取人耳和侧面人脸的关联特征,进行人耳和侧面人脸在特征层的融合.实验结果表明,此方法与单一的人耳或侧面人脸特征识别比较,识别率得到提高.     

一种应用于人脸识别的超分辨率方法

为提高局部保持投影(LPP)在人脸图像超分辨率中的适用性,在LPP中引入典型相关分析(CCA),提出一种相关性增强的局部保持投影方法(CELPP)。CELPP用于提取高分辨率图像与低分辨率图像特征,根据关系学习建立低分辨率图像特征与高分辨率图像特征之间的映射变换,输入低分辨率图像,通过CELPP特征提取和关系映射,得到高分辨率图像,并将其用于人脸识别。对人脸库ORL和Yale进行的实验结果表明,该方法同时考虑了高分辨率图像与低分辨率图像的相似性及同类图像的局部结构性,在基于人脸识别的超分辨率应用中优于LPP和CCA。     

基于深度学习的监督型典型相关分析

典型相关分析（CCA）是利用综合变量对之间的相关关系反映两组指标之间整体相关性的多元统计方法。传统的CCA方法无法有效利用样本的标签信息,导致准确率降低。将类信息融入到深度学习与CCA相结合的深度典型相关分析中,提出一种监督型降维方法DL-SCCA,用于处理带标签的非线性可分数据。在2个独立的深度神经网络（DNN）结构上,增加1个公共的输出维数与数据集类别数相同的全连接层,并且以softmax函数作为该层的激活函数,输出带有概率意义的编码向量。在此基础上,利用全连接输出与样本标签信息之间的交叉熵对DNN进行训练,获得分类性能较优的低维特征。实验结果表明,该方法采用最近邻分类器和网络本身结构得到的分类准确率分别为98.00%和97.82%,相比CCA、DisCCA、DCCA等方法,能够有效利用样本的标签信息,并且具有较优的分类性能。     

一种基于组策略的过滤式特征选择算法

MRMR算法具有快速、高效等优势,在处理高维数据方面较为流行。基于此,提出一种基于组策略的MRMR改进算法（MRMRE）,该算法不仅考虑单个特征属性的相关性与冗余性,同时针对特征组间的相互关系进行研究。算法以MRMR算法为框架,以CCA作为度量基准,选择SVMs作为基分类器,使其特征选择效果提升。在UCI机器学习数据库中图像与基因序列数据集上的大量实验表明：与MRMR算法相比,所提出的算法其特征选择结果具有更高的结果稳定性与分类精度。     

Porous ceramics with tailored pore size and morphology as substrates for coral larval settlement

The growing demand for stony corals as ornamental aquarium animals requires defined aquacultural breeding strategies. For the sexual propagation of corals, material substrates are needed, that attract larvae and support their settlement and development. In this study, five types of highly porous ceramic materials were developed following the example of coral skeleton. The applicability of these settlement substrates was tested using larvae of the stony coral Pocillopora damicornis. Partial sintering of pressed clay pellets, freeze casting of clay and alumina-mullite based slurries and direct foaming of high alkane phase emulsified suspensions (HAPES) using alumina were employed. By the addition of mm-sized spherical polystyrene beads as sacrificial templates during freeze casting (alumina-mullite), superficial pores in the size of the larvae were created. The inorganic substrates featured open porosities between 35% (pressed clay) and 83% (foamed alumina), pore sizes ranging from nm to mm-scale and pore morphologies dominated by interparticle porosity (pressed), lamellar pores (freeze casting) and cellular pore types (direct foaming). The ceramic substrates were incubated in artificial sea water for 3 months to induce necessary biofilm formation and algae growth. Afterwards, individual substrates were exposed to 5 coral larvae, and their settlement behavior was monitored over 14 days. At the end of this period, all ceramic materials were successfully accepted as settlement substrates, with a mean settlement rate of 46.2%, and no significant differences between the substrate types. On samples with large surface superficial pores, a significantly reduced survival of settled larvae (79%) compared to the other porous materials (93–98%) was determined, suggesting a non-ideal surface topography. While alumina foam samples (HAPES) exhibit the most promising results in terms of settlement and survival of larvae, clay-based substrates provide a more economic solution for the sexual propagation of corals in aquaculture.     

A framework for joint estimation of age,gender and ethnicity on a large database

Human age, gender and ethnicity are valuable demographic characteristics. They are also important soft biometric traits useful for human identification or verification. We present a framework that can estimate the three traits jointly. The joint estimation framework could deal with the mutual influence of age, gender, and ethnicity implicitly. Under this joint estimation framework, we explore different methods for simultaneous estimation of age, gender, and ethnicity. The canonical correlation analysis (CCA) based methods, and partial least squares (PLS) models are explored under our joint estimation framework. Both the linear and nonlinear methods are investigated to measure the performance. We also validate some extensions of these methods, such as the least squares formulations of the CCA methods. We found some consistent ranking of these methods under our joint estimation framework. More importantly, we found that the CCA based methods can derive an extremely low dimensionality in estimating age, gender and ethnicity. An analysis of this property is given based on the rank theory. The experiments are conducted on a very large database containing more than 55,000 face images.     

基于邻域信息的图像分割技术

提出一种基于图像邻域信息的分割方法.首先,根据像素点邻域信息得到高维特征向量;然后采用典型相关分析(CCA)改进线性判别分析(LDA)中的变换矩阵,使得特征向量的降维具有自适应性;最后用最近邻法对降维后的特征向量进行分类,从而实现了图像的分割.试验中,选取人脸图像分割来验证该方法,结果显示出其具有良好的分类效果.