码分复用数字音频水印嵌入方案

结合人耳听觉特性，用扩频方法在音频信号中的听觉不敏感区域嵌入多路数字水印，从而实现码分复用．在对嵌入信息的检测中，除了少量系统参数外，数字水印的提取不需要原始信号，并且毋需精确同步．实验表明，该水印具有很好的隐蔽性；尤其是对常见攻击，MP3编码具有良好的稳健性．     

Recent developments in the statistical processing of textual data

Statisticians are accustomed to processing numerical, ordinal or nominal data. In many circumstances, such as socio-economic, epidemiologic sample surveys and documentary data bases, this data is juxtaposed with textual data (for example, responses to open questions in surveys). This article presents a series of language-independent procedures based upon applying multivariate techniques (such as correspondence analysis and clustering) to sets of generalized lexical profiles. The generalized lexical profile of a text is a vector whose components are the frequencies of each word (graphical form) or ‘repeated segment’ (sequence of words appearing with a significant frequency in the text). The processing of such large (and often sparse) vectors and matrices requires special algorithms. The main outputs are the following: (1) printouts of the characteristic words and characteristic responses for each category of respondent (these categories are generally derived from available nominal variables); (2) graphical displays of the proximities between words or segments and categories of respondents; (3) when analysing a combination of several texts: graphical displays of proximities between words or segments and each text, or between words or segments and groupings of texts. The systematic use of ‘repeated segments’ provides a valuable help in interpreting the results from a semantic point of view.     

用于语音识别中的SOFM矢量量化方法

该文讨论了神经网络语音识别系统中的ＫｏｈｏｎｅｎＳＯＦＭ神经网络的矢量量化（ＶＱ）的原理与过程,着重解决了其在实际应用时的若干问题,包括网络大小,学习步幅函数以及矢量量化过程中的公平竞争学习的控制函数的选定。过讨论了基于听觉模型的声学特性的提取与处理,这与ＶＱ性能息息相关。系统应用的实验表明,该神经网络在语音识别系统中具有良好性能。     

非晶高聚物对聚(ε-己内酯)在其与苯乙烯-丙烯腈共聚物相容共混物中球晶生长速率的影响

本文研究了聚(ε-己内酯)(PCL)在其与苯乙烯-丙烯腈共聚物(SAN)的相容共混物中球晶生长速率与共混组成和结晶温度的关系.发现聚己内酯的球晶生长速率随着SAN的含量增加而下降.由于PCL与SAN是相容共混物,因此在用二次成核动力学方程描述PCL球晶生长速率时,我们引进了相互作用参数X.结果由共混体系的结晶动力学方程计算到的X值与由平衡熔点下降方法计算到的X_(23)值是相同的;而PCL晶体的折叠表面自由能则随着SAN含量的增加而下降.这些结果说明非晶高聚物SAN有碍于PCL球晶的生长.     

脑干听觉诱发电位波形分析建模、验证与算法优化

脑干听觉诱发电位是声觉器官受刺激后产生的一种生物电，反映了相应感觉通路及神经元群的活动，可用作听力阈值的客观判断，也可对多种耳疾和脑部病变作出诊断，因而成为当今临床医学和生物医学信号处理中的一个重要课题．文中介绍了脑干听觉诱发电位潜伏期（Ｌ）与刺激声强（Ｉ）之间关系曲线的作用，及其拟合建模和回归算法优化，还用实测数据对选定的拟合曲线模型进行了回归方差分析验证．     

江油附子茎段和茎尖植株再生技术的初步研究

选择生长健壮、无病虫害的附子茎段和顶芽为外植体,接种到附加不同植物生长素的培养基上进行离体培养研究.结果表明:MS+6-BA 2 mg.L-1+NAA 0.3 mg.L-1的培养基诱导无菌苗效果最佳,1/2MS+IBA 1 mg.L-1+KT 0.5 mg.L-1生根培养基,生根率达80%.利用该技术可以成功的建立附子再生植株的实验体系.     

A novel audio watermarking scheme using multiscale wavelet modulation

A novel audio watermarking scheme to embed robust and inaudible watermarks for the purpose of copyright protection is proposed. The key innovation is to add time-frequency redundancy into watermark signals by multiscale wavelet modulation. In order to maximize the watermarking strength within perceptual constraints, the signals synthesized from different scales are masked using a frequency auditory model, respectively, and then intergrated to form the final watermark signal. The detection structure is built using the redundancy in watermark signals, and the performance is further enhanced by modeling the statistical behaviors of wavelet coefficients as generalized Gaussian distribution. The use of original audio signal is not required in watermark detection. The experimental results show that our approach can achieve not only good transparency but also satisfying robustness to common audio manipulations.     

基于A*改进算法的机械臂避障路径规划

为解决机械臂在运行时可能与工作空间中的障碍物发生碰撞,提出圆柱体包络碰撞检测法和基于A*算法的变步长分段搜索法.运用圆柱体包络碰撞检测法,将碰撞检测问题转化为计算圆柱体间的位置关系.变步长分段搜索法解决了应用传统A*算法可能出现的搜索数据量大甚至搜索死循环等问题.分别在基于OpenGL的三维仿真实验平台和实际系统中验证了该方法的有效性和可行性.     

基于听觉场景分析的近讲语音增强算法

针对近讲场景,提出一种双麦克近讲语音增强算法。该算法基于耳间延时差(ITD)、耳间强度差(IID)特征来区分目标语音,利用声学掩蔽效应,实现目标语音对背景噪声的分离降噪。与传统的单麦克增强方法相比,该算法可消除多种类型噪声且对语音造成的损伤较小。实验表明:该算法能将8～33dB的白噪声、音乐噪声、广播噪声3种噪声类型的带噪语音的信噪比提高到36dB以上,同时保持较高的目标语音可懂度。对于冲击噪声的带噪语音也具有较好的降噪效果。     

多段折线逼近法在智能气压传感器设计中的应用

提出了在智能气压传感器算法设计中,利用多段折线逼近法实现非线性校正的一种应用技巧,并以计算出的结果与实验标定数据相比较,验证了该方法的可行性.