Face detection using discriminating feature analysis and Support Vector Machine

This paper presents a novel face detection method by applying discriminating feature analysis (DFA) and support vector machine (SVM). The novelty of our DFA-SVM method comes from the integration of DFA, face class modeling, and SVM for face detection. First, DFA derives a discriminating feature vector by combining the input image, its 1-D Haar wavelet representation, and its amplitude projections. While the Haar wavelets produce an effective representation for object detection, the amplitude projections capture the vertical symmetric distributions and the horizontal characteristics of human face images. Second, face class modeling estimates the probability density function of the face class and defines a distribution-based measure for face and nonface classification. The distribution-based measure thus separates the input patterns into three classes: the face class (patterns close to the face class), the nonface class (patterns far away from the face class), and the undecided class (patterns neither close to nor far away from the face class). Finally, SVM together with the distribution-based measure classifies the patterns in the undecided class into either the face class or the nonface class. Experiments using images from the MIT-CMU test sets demonstrate the feasibility of our new face detection method. In particular, when using 92 images (containing 282 faces) from the MIT-CMU test sets, our DFA-SVM method achieves 98.2% correct face detection rate with two false detections.     

基于网络状态的入侵检测模型

本文提出了一种基于网络状态的入侵检测模型。该模型结合网络入侵的目标和特点，利用有穷自动机理论，基于网络协议来实现进程和操作系统的状态建立，从而可以发现未知的入侵，本文论证了应用该模型的可靠性，并利用通用入侵检测框架CIDF对应用该模型的入侵检测系统进行了描述，最后与其它入侵检测模型进行了比较。     

An application of design space for assembly process reasoning to utilize current assembly plant resources for new product family members

A successful and profitable product platform strategy requires both product family architecture and assembly process reasoning. New product family member production cost and time can be significantly reduced by utilizing available assembly resources, which can be achieved through systematic assembly process reasoning. A method to utilize existing assembly plant resources, during the development of new product family members, requires comparing feasible assembly processes with exiting assembly plants. The set of feasible assembly sequences for a product family member is modelled by developing an assembly sequence design space, which is combinatorial in nature, and applying constraints on the space. Models that capture effects of constraints on these spaces, explicitly represent feasible regions, and efficiently enumerate designs within this space are investigated. The feasible space is then searched to determine new product assembly sequence that will require minimum change in the current assembly plant. An automotive front structure family is utilized to demonstrate application of the assembly sequence space to perform assembly reasoning to increase exiting assembly plant resource utilization.     

电子鼻在天然苹果香精检测标准建立中的应用研究

利用电子鼻检测仅知生产日期的天然苹果香精样品，并对所获数据进行PCA及DFA分析，筛选信息重复的样品。结果表明：生产日期与产品质量有着密切的联系；通过筛选，参与建立天然苹果香精检测标准的样品只需20个，比原来的34个样品减少了14个。     

Cloning of inulin fructotransferase (DFA III-producing) gene from Arthrobacter globiformis C11-1

A gene encoding an inulin fructotransferase (DFA III-producing) [EC 2.4.1.93] from Arthrobacter globiformis C11-1 was cloned and the nucleotide sequence was determined. The cloned fragment contained a 1353 bp open reading frame. The initiation codon was estimated to be an unusual codon, GTG. The gene encoded a signal peptide (40 amino acid residues) for secretion. The molecular mass of the native enzyme was calculated as 43,400 Da from the sequencing data. The deduced amino acid sequence of the enzyme had 74.0 % homology with that of inulin fructotransferase (DFA III-producing) from Arthrobacter sp. H65-7. It also had 45.1% homology with that of inulin fructotransferase (DFA I-producing) [EC 2.4.1.200] from Arthrobacter globiformis S14-3. The enzyme produced in the culture supernatant of an Escherichia coli clone was purified to the electrophoretically homogeneous stage. The N-terminal amino acid sequence of the cloned enzyme secreted in the broth was the same as that of the native enzyme from A. globiformis C11-1. Therefore, on this enzyme, it is estimated that the cleavage sites by the signal peptidase for secretion of A. globiformis C11-1 and E. coli JM109 are the same.     

面向并行工程的DFA理论与实践

在产品的生产过程中,大约有1/3左右的人力在从事有关产品装配的活动。G.Hird等人指出,在工业化国家中,超过40％以上的生产费用用于产品的装配。因此,降低装配费用,历来是企业界人士和专家们关注的焦点之一。 产品的可装配设计(DFA-Design for Assembly)正是为了降低装配费用而成长起来的高新技术之一。本丈论述了DFA的基本概念、产生背景、发展方向及实现策略。     

SIP代理服务器抗拒绝服务攻击自防御模型

通过对拒绝服务攻击的原理、类型、特征及SIP网络面对的典型DoS攻击的分析,结合Client Puzzle思想、DFA、远程联动、本机联动技术及SIP网络特点、SIP代理服务器工作原理,提出了SIP代理服务器抗拒绝服务攻击自防御模型,并设计相应的实验环境对该模型抵御DoS攻击的能力进行测试,实验结果证明该模型在增加代理服务器转发信令毫秒级的时间延迟下,可有效地抵御针对SIP代理服务器发起的DoS攻击,增强SIP系统服务的可用性.     

面向装配设计过程中的优化方法

面向装配设计过程中往往会涉及到几何形状的复杂性和任意性 ,以及大量的非线性数值约束和大量的不可形式化的约束 ,因此需要有效的优化方法 ,本文着重介绍了针对装配过程中装配序列优化、装配公差分析及可装配性评价等各个功能模块优化求解算法 ,讨论了其存在的问题及发展趋势。     

在线用户打分行为长记忆效应与信任关系研究

对在线打分行为的动态研究能够帮助深入理解社交网络用户集群行为和信任关系的演化机制,当前许多在线系统用户能够通过对物品进行打分传达自己的观点。通过去趋势波动分析法研究了用户打分行为在信任关系建立前后的长记忆效应,并通过随机化打分时间和信任时间建立零模型,最后进行用户打分行为异质性分析。采用Epinions数据集进行实证研究,结果表明用户打分的长记忆效应在信任关系建立前出现下降趋势（8.06%）,并于之后逐步回升（8.43%）,而在两个零模型中赫斯特指数分别稳定在0.5和0.6左右,且用户长记忆效应变动与用户度呈正相关,Pearson相关系数分别为0.9358和0.9278。该工作有助于深入理解用户集群行为和信任关系的动态演化机制。     

An investigation into design and manufacturing of mechanical conveyors systems for food processing

This paper presents the results of a research investigation undertaken to develop methodologies and techniques that will reduce the cost and time of the design, manufacturing and assembly of mechanical conveyor systems used in the food and beverage industry. The improved methodology for design and production of conveyor components is based on the minimisation of materials, parts and costs, using the rules of design for manufacture and design for assembly. Results obtained on a test conveyor system verify the benefits of using the improved techniques. The overall material cost was reduced by 19% and the overall assembly cost was reduced by 20% compared to conventional methods.