The scaling problem in the pattern recognition approach to machine translation

Statistical machine translation (SMT) has proven to be an interesting pattern recognition framework for automatically building machine translations systems from available parallel corpora. In the last few years, research in SMT has been characterized by two significant advances. First, the popularization of the so called phrase-based statistical translation models, which allows to incorporate local contextual information to the translation models. Second, the availability of larger and larger parallel corpora, which are composed of millions of sentence pairs, and tens of millions of running words. Since phrase-based models basically consists in statistical dictionaries of phrase pairs, their estimation from very large corpora is a very costly task that yields a huge number of parameters which are to be stored in memory. The handling of millions of model parameters and a similar number of training samples have become a bottleneck in the field of SMT, as well as in other well-known pattern recognition tasks such as speech recognition or handwritten recognition, just to name a few. In this paper, we propose a general framework that deals with the scaling problem in SMT without introducing significant time overhead by means of the combination of different scaling techniques. This new framework is based on the use of counts instead of probabilities, and on the concept of cache memory.     

基于极坐标区间运算的2D形状匹配

形状匹配是遥感图像目标识别、字符识别、手形识别和步态识别等任务中的关键步骤之一.针对刚体识别任务中形状匹配易受方向、尺度和位置等仿射变化量影响的情况,提出了一种新的基于极坐标区间运算的2D形状匹配算法.该算法首先以形状区域的中心点为极点,区域的最长轴方向为极轴,对形状区域进行归一化的极坐标变换;然后定义了同一角度对应的区域内点区间之间的运算;最后定义了两个区域归一化极坐标变换结果在区间运算下的相似度函数,用以表征两个区域之间的匹配度.从可见光遥感图像中提取的实物图像实验结果证明,该方法能够有效归类相似形状,并能区分各类不同的形状.     

图像的平视化处理

摄像机拍摄图像时,会使物体在同一水平面上的边缘呈现曲线形状,给图像处理方法的实现和图像边缘位置的定位带来不便。这篇文章研究了把图像的曲线边缘变为直线边缘的图像平视化处理方法,推导了像素点和空间点的对应关系公式,陈述了图像平视化处理的思路,并对现场采集的图像进行了平视化处理。通过试验,这篇文章的平视化方法取得了较好的效果。     

基于英汉平行语料库术语词典的自动抽取

该文提出了一种从英汉平行语料库中自动抽取术语词典的算法。采用的是已对齐好的双语语料,中文经过了分词处理。利用英文和中文词性标注工具对英文语料和中文语料分辨进行词性标注。统计双语语料库中的名词和名词短语生成候选术集。然后对每个英文候选术语计算与其相关的中文翻译间的翻译概率。再通过设定阈值过滤掉一些与该英文候选词无关的中文翻译,最后通过贪心算法选取概率最大的词作为该英文候选词的中文翻译。     

平移不变小波去噪法在胎儿心率提取中的应用

用传统的小波阈值法对超声多普勒胎儿心脏回波信号进行去噪处理时会产生Psuedo Gibbs现象,影响了去噪效果和胎儿心率值的提取。为了解决这个问题,在此基础上提出了平移不变小波法去噪。实验结果表明,该方法有效的抑制了Psuedo Gibbs现象,剔除了信号中的噪声,识别出被噪声淹没的有用信号,有利于心率值的提取。     

UML2.0顺序图的形式化研究

在UML2.0规范中顺序图的语义仍然是以自然语言的形式描述的,为实现对顺序图的自动化分析和验证,必须为顺序图定义一种形式化的语义模型.为此首先给出了UML顺序图的一种符合BNF范式的形式化语法,然后为该语法中的非终止符定义转换规则,将UML顺序图中的基本动作转换为加标Petri网组件,最后定义了各种合成操作,利用这些合成操作可以将UML顺序图的加标Petri网组件转换为加标Petni网.     

利用语义词典Web挖掘语言模型的无指导译文消歧

为了解决困扰词义及译文消歧的数据稀疏及知识获取问题,提出一种基于Web利用n-gram统计语言模型进行消歧的方法.在提出词汇语义与其n-gram语言模型存在对应关系假设的基础上,首先利用Hownet建立中文歧义词的英文译文与知网DEF的对应关系并得到该DEF下的词汇集合,然后通过搜索引擎在Web上搜索,并以此计算不同DEF中词汇n-gram出现的概率,然后进行消歧决策.在国际语义评测SemEval-2007中的Multilingual Chinese English Lexical Sample Task测试集上的测试表明,该方法的P_mar值为55.9%,比其上该任务参评最好的无指导系统性能高出12.8%.     

一种基于同步树替换文法的统计机器翻译模型

提出一种基于同步树替换文法的机器翻译模型.相对于基于短语的模型,此模型可以对远距离结构性调序和非连续短语翻译进行建模;相对于基于同步上下文无关文法模型,此模型可以对任何层次上的树节点调序进行建模.因此,该模型可以为处理语言结构间的异构对应问题提供有效的解决途径. 在两组风格差异较大的数据集上进行的实验均验证了基于同步树替换文法的模型相对于基于短语模型和基于同步上下文无关文法模型的稳定优势.     

实时图像增强算法改进及FPGA实现

针对复杂背景的多目标图像,提出了一种基于直方图的实时自适应图像增强方法.该方法根据自适应直方图窗口选择高低阈值,通过灰度线性变换及灰度级等间距密度均衡进行图像增强.利用该算法增强图像视频时,采用FPGA,通过并行处理结构及流水线技术,可实时处理每秒50帧780×582×12bits的可见光图像.在处理视频的过程中,由前一帧图像的直方图信息,来增强后一帧图像.理论分析和实验结果均表明,该算法克服了直方图均衡及平台直方图均衡增强图像引起的灰度断层现象,有效地增强了图像的对比度,提高了图像质量.     

Auto-generation of IEC standard PLC code using t-MPSG

The objective of this paper is to reduce the development time of a PLC (Programmable Logic Controller) by automating the task of code generation. For this purpose, we applied t-MPSG (Timed-Message Based Part State Graph). The t-MPSG is an extended finite state automata used to model and generate an execution module for a real-time shop floor controller system. In our proposed method, t-MPSG is used to model the formal specification of the controller system that can be translated into textual structure. After the verification of the t-MPSG model, it can be used as an input to the plc-builder tool. The plc-builder tool is an extended version of a conventional MPSG simulator. It can be used to translate the textual structure of the t-MPSG into an IEC standard PLC code. Finally, the generated code can be downloaded to a PLC emulator or a PLC device for the purpose of simulation and execution. The similarity in the hierarchical structure of the t-MPSG and the IEC standard PLC program has made it convenient to transform from one form to another. Furthermore, an illustration of the methodology to auto-generate IEC standard PLC code using t-MPSG is explained with a suitable example. Recommended by Editorial Board member Young Soo Suh under the direction of Editor Jae Weon Choi. This work was partially supported by Defense Acquisition Program Administration and Agency for Defense Development under the contract (UD080042AD). Devinder Thapa is a Postdoc Research Fellow in the Department of Industrial & information systems at Ajou University, Korea. He completed his Ph.D. from Ajou University in Industrial and Information Systems Engineering. His area of research is related to manufacturing automation and intelligent decision support systems. Chang Mok Park is a Professor in the Department of Technology & Systems Management at Induk Institute of Technology. He completed his Ph.D. in 2002 from Ajou University in Industrial Engineering. His research interest is related to manufacturing optimization, discrete event system simulation and signal analysis. Sang C. Park is an Associate Professor in the Department of Industrial & Information Systems Engineering at Ajou University. He received his B.S., M.S., and Ph.D. degrees from KAIST in 1994, 1996, and 2000, respectively, all in Industrial Engineering. His research interests include geometric algorithms in CAD/CAM, process planning, engineering knowledge management, and discrete event system simulation. Gi-Nam Wang is the Head and a Professor in the Department of Industrial & Information Systems Engineering at Ajou University, Korea. He completed his Ph.D. in 1992 from Texas A&M University, in Industrial Engineering. He has worked as Visiting Professor at University of Texas at Austin during 2000–2001. His area of research is related to Intelligent Information & manufacturing systems, system integration & automation, e-Business solutions and image processing.