多特征融合的鸟类物种识别方法

针对短时窗平均/长时窗平均算法从次声台站监测数据中提取的信号仍然包含噪声的问题,对支持向量机和人工神经网络的机器学习方法进行了研究。采用小波包分解的方法对信号进行重构,提取出各频带内的重构信号能量特征,对事件信号和噪声进行了识别实验,并分析了提高识别能力的方法,为工程应用提供理论参考。实验结果表明,在训练数据集不大的情况下,通过优化模型结构可以将两种方法的识别能力提高到可以接受的水平。     

Heuristic Backtrack Algorithm for Structural Match and Its Applications

The concept WALKING on structures is proposed, and the partial ordering between a structure and a query structure (substructure) is also created by means of WALKING. Based upon the above concepts, authors create the Heuristic-Backtracking Algorithm (HBA) of structural match with high performance. In the last part of the paper, the applications of HBA in molecular graphics, synthetic planning, spectrum simulation , the representation and recognition of general structures are discussed.     

Evidence for strong heterodimeric interactions of phenylboronic acids with amino acids

Strong heterodimeric interactions of phenylboronic acids with l-proline or betaine are evident in the solid state. The interaction energy is over 23 kcal/mol (at MP2/6-31+G^∗).     

联合击中击不中变换相关器

将联合击中击 中变换相关器用于二值图像的目标识别。与常用的联合变换相关器相比，联合击中击不中变换相关器对联合变换相关器输入图像和参考图像进行互补编码。编码的目的在于通过联合变换相关器来一步实现形态学中的击中击不中变换。     

Post Recognition Speech Localization

How much does knowledge regarding a certain spoken word or phrase help with its localization? This is a very fundamental question for speech processing, and will be partially addressed in this paper. In particular, this work will utilize prior information regarding the contents of a speech signal in order to improve the artificial localization of it using Time delay of arrival (TDOA) between two microphones. The prior information, which is used to develop a very simple frequency-selective phase transform (FPT), increases the effective SNR by only using a subset of the highest SNR frequencies in the Phase Transform. Simulations in a reverberant environment show that the proposed approach can more robustly and accurately localize speech sources. For 20 ms signal segments, it is shown that using a subset of 45 percent of available speech frequency bins is superior to using 30, 60, or 100, where using 100 corresponds to the standard Phase Transform.     

Light power density distribution of image spot of laser triangulation measuring

HaiMing Wang [Wang HaiMing, Long range optical triangulation utilising collimated probe beam. Optics and Lasers in Engineering, 23(11), 1995, 41–52] studied the influences of the size of the object spot on the image spot in laser triangulation measuring. This paper points out that the study method of HaiMing Wang is inappropriate from two points of view. The size and relative power density distribution of the image spot can be calculated either when assuming the object spot is an ideal point, which emits a uniform spherical wave over the object, or when assuming that the object surface is a light scattering surface. Both the conditions show that the size of the image spot is much bigger than that of HaiMing Wang’s.     

一种提高光学目标识别速度的多通道方案

本文以形态学击中击不中变换作为理论依据，以非相干光相关器作为光学实现硬件，推导了在非共轴相关条件下相关点位置的计算公式。基于系统的平移不变性，在无需增加任何分束器的情况下实现了多通道并行处理方案，不仅充分利用了空间光调制器的工作幅面，而且大大提高了识别速度，体现了光学方法的并行性和快速性。文中给出了用四通道结构，以正面人脸图像作为测试对象的实验结果。     

Characterization of fatty oils by pattern recognition of triglycerides profiles

The triglyceride profiles for corn, soybean, sunflower, grapestone, olive and olive husks oils were obtained by high temperature capillary gas chromatography on HT-FS capillary columns coated with OV-17-OH. The data were treated by methods of computerized pattern analysis, involving principal component analysis, discriminant analysis, and hierarchical clustering. A graphical representation of results as “star symbol plots” allows prompt visual recognition of the characteristic profiles and straightforward qualitative identification of unknowns as well as recognition of adulterations.     

汉语耳语音孤立字识别研究

耳语音识别有着广泛的应用前景,是一个全新的课题.但是由于耳语音本身的特点,如声级低、没有基频等,给耳语音识别研究带来了困难.本文根据耳语音信号发音模型,结合耳语音的声学特性,建立了一个汉语耳语音孤立字识别系统.由于耳语音信噪比低,必须对其进行语音增强处理,同时在识别系统中应用声调信息提高了识别性能.实验结果说明了MFCC结合幅值包络可作为汉语耳语音自动识别的特征参数,在小字库内用HMM模型识别得出的识别率为90.4%.