A new approach to feature selection based on the Karhunen-Loeve expansion

After surveying existing feature selection procedures based upon the Karhunen-Loeve (K-L) expansion, the paper describes a new K-L technique that overcomes some of the limitations of the earlier procedures. The new method takes into account information on both the class variances and means, but lays particular emphasis on the classification potential of the latter. The results of a series of experiments concerned with the classification of real vector-electrocardiogram and artificially generated data demonstrate the advantages of the new method. They suggest that it is particularly useful for pattern recognition when combined with classification procedures based upon discriminant functions obtained by recursive least squares analysis.     

Greylevel edge thinning: A new method

A comparison is made of the performance of two greylevel thinning algorithm for simple images which contain feature primitives from which more complexes scenes are built. A quantitative assessment of performance is made and a new method proposed.     

Independent component analysis in a local facial residue space for face recognition

In this paper, we propose an Independent Component Analysis (ICA) based face recognition algorithm, which is robust to illumination and pose variation. Generally, it is well known that the first few eigenfaces represent illumination variation rather than identity. Most Principal Component Analysis (PCA) based methods have overcome illumination variation by discarding the projection to a few leading eigenfaces. The space spanned after removing a few leading eigenfaces is called the “residual face space”. We found that ICA in the residual face space provides more efficient encoding in terms of redundancy reduction and robustness to pose variation as well as illumination variation, owing to its ability to represent non-Gaussian statistics. Moreover, a face image is separated into several facial components, local spaces, and each local space is represented by the ICA bases (independent components) of its corresponding residual space. The statistical models of face images in local spaces are relatively simple and facilitate classification by a linear encoding. Various experimental results show that the accuracy of face recognition is significantly improved by the proposed method under large illumination and pose variations.     

Introduction to the Special Issue on Biometrics: Progress and Directions

The guest editors provide an overview of the articles selected for this special issue. The issue's goal is to document the current state-of-the-art, acknowledge the latest breakthroughs achieved by scientists working in the area of biometric recognition, and identify future promising research areas. It is thought the selection of papers discussed should give readers a good idea of where researchers have been focusing, both on long- studied problems still needing more work and on newer challenges. A fundamental of the field of biometrics is an ever-increasing need for better recognition and stronger security. But, as public and commercial biometric deployments increase in number, there is also more need to understand privacy issues and to provide greater ease-of-use. The volume and quality of papers in this special issue indicate that much progress has been made in many aspects of the biometrics field and that there are challenging and promising future directions still to follow.     

一种模块化2DPCA和CSLDA相结合的人脸验证算法

在CSLDA方法的基础上进行改进,和模块化2DPCA相结合,提出了一种模块化2DPCA+CSLDA的人脸验证方法.CSLDA将图像矩阵转化为向量进行处理,数据维数很大,计算复杂,对图像整体处理没有考虑到图像的局部特征.针对这些缺点,新方法从原始数据出发,对二维数据进行分块后采用2DPCA进行特征抽取,能有效抽取图像的局部特征,得到替代原始图像的低维的新模式.然后对新模式施行CSLDA,即基于客户相关子空间的线性判别分析方法,不仅考虑到类内、类间的差异,弥补了PCA的缺陷;而且客户相关(CS)子空间可以较好地描述不同个体人脸之间的差异性,比传统的个体特征脸具有更好的判别能力.在XM2VTS人脸库上按照Lausanne协议和ORL库上对原CSLDA和新方法进行评价和测试的结果表明, 新方法在验证效果上优于CSLDA方法.     

Availability of luminous flux below a bended light-pipe: Design modelling under optimal daylight conditions

To quantify an averaged directionality of light beams transmitted by hollow light guides the asymmetry parameter has been introduced which is computed from the luminous intensity solid or curve. In many situations, the light guides with upper part elbowed toward the Southern sky are able to collect more light than a corresponding straight pipe. However, under different sunlight and skylight conditions the internal reflections are more complex, resulting in a large variety of light-pipe efficiency. A more realistic prediction of the luminous intensity solid under manifold exterior daylight conditions is significant for the optimal design and installation of passive illumination systems using bended light guides. Here, the orientation of the elbow is crucial for an efficient utilization of daylight, especially during a clear day. This study documents that the maximum efficiency of the bended light-pipe is reached when its upper part is pointed nearby the sun position but directly towards the sun.     

XBIC/μ-XRF/μ-XAS analysis of metals precipitation in block-cast solar silicon

The results of the investigations of the interaction between the different impurities in intentionally contaminated block-cast multi-crystalline silicon by means of synchrotron-based microprobe techniques XBIC (X-ray beam induced current), μ-XRF (X-ray fluorescence microscopy) and μ-XAS (X-ray absorption microspectroscopy) recently implemented at beamlines ID-21 and ID-22 of ESRF, Grenoble, are presented. It was found that Si₃N₄/SiC particles frequently observed in the upper part of multi-crystalline Si blocks represent effective sinks for Fe and Cu impurities. The amount of precipitated iron was the same order magnitude both at nitride and carbide particles. The amount of Cu precipitated at the SiC inclusions was significantly larger than that at Si₃N₄ rods. Chemical state of the copper precipitates was identified as copper-rich silicide Cu₃Si. The anneal at 950 °C that is known to enhance oxygen precipitation in silicon was found to accompany with the enhanced formation of nanoscale iron disilicide precipitates both inside the grains and at grain boundaries.     

Radiance distribution on densely overcast skies: comparison with cie luminance standard

In this contribution some data measured in Europe are evaluated to enable the comparison of real radiance distributions with the CIE (International Commission on Illumination) luminance gradation approximation of the densely overcast sky. A question is posed whether such a model sky exists in reality, although the frequency of its occurrence might be different in different seasons and in regions with different climates. Eleven cases are evaluated in Figs. 1–3 with the comparison to the cosine and the exponential approximation curves. It seems that the measurement data agree with the CIE Standard but further tests and results obtained during the International Daylight Measurement Programme have to gather more evidence.     

Low threading dislocation density Ge deposited on Si (1 0 0) using RPCVD

Epitaxial Ge layer growth of low threading dislocation density (TDD) and low surface roughness on Si (1 0 0) surface is investigated using a single wafer reduced pressure chemical vapor deposition (RPCVD) system. Thin seed Ge layer is deposited at 300 °C at first to form two-dimensional Ge surface followed by thick Ge growth at 550 °C. Root mean square of roughness (RMS) of ∼0.45 nm is achieved. As-deposited Ge layers show high TDD of e.g. ∼4 × 10⁸ cm⁻² for a 4.7 μm thick Ge layer thickness. The TDD is decreasing with increasing Ge thickness. By applying a postannealing process at 800 °C, the TDD is decreased by one order of magnitude. By introducing several cycle of annealing during the Ge growth interrupting the Ge deposition, TDD as low as ∼7 × 10⁵ cm⁻² is achieved for 4.7 μm Ge thick layer. Surface roughness of the Ge sample with the cyclic annealing process is in the same level as without annealing process (RMS of ∼0.44 nm). The Ge layers are tensile strained as a result of a higher thermal expansion coefficient of Ge compared to Si in the cooling process down to room temperature. Enhanced Si diffusion was observed for annealed Ge samples. Direct band-to-band luminescence of the Ge layer grown on Si is demonstrated.     

Silicon solar cells with antireflection diamond-like carbon and silicon carbide films

Optical properties of diamond-like carbon and silicon carbide (SiC) films in dependence on deposition conditions were investigated. It was established that the films having refractive index from 1.6 to 2.3 may be obtained. The film optical bandgap and hardness may be changed from 1.5 to 4 eV and from 1 to 20 GPa, correspondingly. The films were deposited onto the front side of silicon solar cells (SCs). It has been shown that deposition of single- or two-layer diamond-like carbon antireflection (AR) coatings enables the SCs efficiency to be improved 1.35–1.5 times. The improvement is connected with decreasing of reflection losses and passivation of recombination active centers. SiC AR coatings improve the solar cell efficiency up to 1.3 times.