data transformation and attribute subset selection: do they help make differences in software failure prediction?

IEEE, IEEE Comp Soc, Tech Council Software Engn

Image Ratio Features for Facial Expression Recognition Application

Video-based facial expression recognition is a challenging problem in computer vision and human-computer interaction. To target this problem, texture features have been extracted and widely used, because they can capture image intensity changes raised by skin deformation. However, existing texture features encounter problems with albedo and lighting variations. To solve both problems, we propose a new texture feature called image ratio features. Compared with previously proposed texture features, e. g., high gradient component features, image ratio features are more robust to albedo and lighting variations. In addition, to further improve facial expression recognition accuracy based on image ratio features, we combine image ratio features with facial animation parameters (FAPs), which describe the geometric motions of facial feature points. The performance evaluation is based on the Carnegie Mellon University Cohn-Kanade database, our own database, and the Japanese Female Facial Expression database. Experimental results show that the proposed image ratio feature is more robust to albedo and lighting variations, and the combination of image ratio features and FAPs outperforms each feature alone. In addition, we study asymmetric facial expressions based on our own facial expression database and demonstrate the superior performance of our combined expression recognition system.

Petri网理论在柔性制造系统(FMS)性能分析中的应用

本文用 Petri 网的一个子类——时间事件图对流水车间型和作业车间型的柔性制造系统(FMS)建模并进行理论分析,给出了可行排序的判定条件及系统中托盘数量配置与系统生产率的关系,对系统的主要性能指标,如生产周期、工件驻留时间、在制品库存等给出了定量描述.这些结果为系统的设计和运行提供了理论依据.

复杂离散制造过程建模与性能评价方法研究

离散制造系统是制造业中最为复杂的制造系统之一，其生产特点表现在：生产方式的异步、并行；生产柔性强；调度复杂性高等方面。要实现对复杂离散制造系统的控制，首先必须对其建模。随着制造系统自动化程度提高，规模的增大，组成因素的复杂化，性能评价问题变得日益复杂并越来越受到人们的重视。提供直观的模型描述方法、有效的数学理论工具和模型分析方法及分析软件，是制造系统性能评价迫切需要解决的问题。为此本文进行了以下工作： 首先，提出了一种支持性能评价的制造过程模型框架。该框架体现了车间生产闭环控制机制，全面描述了车间制造过程相关组成要素之间的关系，可用于指导制造过程的建模和基于性能评价的优化控制。模型框架核心部分是生产过程模型、过程控制模型和过程性能模型，这三部分构成了制造过程模型。该过程模型在工厂模型的支持下，由事件模型驱动，受策略模型约束，通过一些过程准则和方法进行测量和监视，对性能相关数据进行统计、分析和评估，反映在性能指标上。 第二，确定了离散车间性能评价指标体系。该指标体系建立了不同性能指标与不同层次、不同类型的制造过程以及不同性能评价主体之间的关联，为选择评价方法奠定了基础。该指标体系按照车间目标将性能指标分为三大类，即时间、质量、成本；每一类细化为三个层次，分别是基本性能指标、扩展性能指标和综合性能指标。 第三，基于随机Petri网建立了具有随机机器故障的制造过程性能模型。该模型包含两个基本的状态转换环，将正常状态和异常情况分别描述；基于故障的随机特征，设置了服从泊松分布的故障发生模块；根据不同的故障发现模式，衍生出若干状态转换环，清晰地将故障发生、发现分离；还考虑了中断作业的处理，使模型更接近实际系统动态行为特征。另外，本文设计了考虑成本信息的作业优先级调度规则，并基于随机高级Petri网（SHLPN）应用到混流装配线动态调度问题中；针对流水线上存在的一类共享资源—维修工人，建立了考虑系统状态的维修排序规则SHLPN模型。通过数学分析和仿真方法结合实现了对以上制造过程模型的性能评价，验证了模型的有效性。 第四，针对性能评价的需求和Petri网在制造过程成本分析方面存在的不足，提出了一种适合制造过程成本分析的价格时间Petri网，并给出了相应的状态空间化简方法。通过对装配生产线多目标工人指派问题的求解，验证了模型和方法的有效性。 最后，根据管理控制过程和生产过程的特点以及它们之间的关系，提出了基于进程代数与Petri网的分层生产控制方法。采用Petri网和进程代数相结合的方法将物流和信息流集成在一个模型中，Petri网描述基于状态的与生产过程有关的过程流，进程代数描述计划、调度等与控制有关的推理过程，构成了两层分布式的车间生产控制结构。 综上可见，本文提出的制造过程建模、性能评价和控制方法，可以支持复杂离散制造系统的分析和综合；基于过程模型实现了复杂离散制造系统的生产过程模拟和性能分析；能够发现影响系统性能的瓶颈环节；为制造过程的改进和制造系统性能的提高提供了依据；进而实现了制造系统自主闭环自适应的优化控制。本文的研究工作具有重要的理论意义和实际应用价值。

工作群体有效性评价及群体工作设计

This thesis study the problem of work group effectiveness and group job design according to extensive literature investigation and the analysis of realistic background. The whole research consists of four parts: (1) The evaluation of work group effectiveness, the aim is to search to criteria that can describe and analysis work group, and explore the cognitive dimensions of effectiveness of Chinese subjects; (2) The study on the relationship between group job characteristic and effectiveness, the aim is to find the general correlation between work group characteristic and effectiveness, and try to search the most important core variable; (3) The study on the preference for the way of group work; (4) The study of the relationship between group composition and group effectiveness, try to examine how different approaches of personnel selection influence work results. The results indicate: (1) The evaluation of group effectiveness mainly consists of two dimensions: performance and the employee's attitude and feelings toward the group, so we can use these two dimensions and corresponding criteria as the standard of effectiveness evaluation. (2) According to the analysis of related literature, we can determine work group characteristic from five aspects: job design, the interdependence among members, group composition, organizational background, group process. (3) Experimental study find that different group job characteristics have different relationship models with effectiveness criteria. Job design, the interdependence among group members, group composition have significant correlation relationships with two kinds of effectiveness criteria; organizational background mainly has relationships with satisfaction criteria; group process mainly has relationships with performance criteria. (4) The choice for people to select the way of group work has the consistency, that is people prefer to "Self--managed work groups"; but different groups have the difference, the main group dimension is the difference between group members and group leaders. (5) Work groups which were composed accordingly to interpersonal attraction have higher levels of communication, coordination, group cohesion and job satisfaction than ability--based groups based. But the performance evaluation under these two conditions has no difference. The thesis analysis and discuss the research results, also point out several questions that needed to be explored further and the possible research directions in the future. This study has some reference value in group--level performance appraisal and reward design, the content and method of group--level job design, and personnel selection of group, etc.