面向大型数据表的粗分析方法

该文运用粗糙集理论,提出一种提取大型数据表中决策规则的方法。首先根据条件属性及属性值的权重将数据表分解成多个子表,再利用粗分析方法分别对各子表简约,综合每个子表归纳出的具有一定置信度的子规则形成规则集,做出推理和决策。     

噪声图像中边界提取方法的研究

在噪声图像中如何有效的提取边界是图像分析中的难点，常用的边界提取方法往往对噪声对敏感，文章针对这一情况采用了保持图像细节的多级组合滤波器先将噪声从图像中分离出来，然后再对图像进行边界提取，实验表明，该方法可以在有效的抑制噪声的同时提取出有意义的边界。     

Accessing Video Contents through Key Objects over IP

In order to support content-based video database access over the Internet Protocol (IP), achieving the following objectives are important: (i) video query by a representative object (key object) or some statistical characterization of the target contents, (ii) bandwidth-efficient browsing over IP, and (iii) scalable and user-centric video transmission over a heterogeneous and variable-bandwidth network. We present a video object extraction and scalable coding system designed to meet the above objectives. In our system, key objects of meaning to video database users are generated via a human-computer-interaction procedure, and are tracked across frames. Given a key object, an algorithm classifies a subset of its VOPs as key VOPs. This subset forms the basis of a highly bandwidth-efficient base layer for supporting activities such as browsing and refining queries. Over the base layer, a number of enhancement layers can be defined to progressively increase the spatial and temporal resolutions of retrieved video. It is expected that heterogeneous users can subscribe to different numbers of the enhancement layers according to their own conditions, such as access authorization, available connection bandwidth, and quality preference.     

Application of InP Neural Network to ECG BeatClassification

This paper presents an application of a hybrid neural network structure to the classification of the electrocardiogram (ECG) beats. Three different feature extraction methods are comparatively examined: discrete cosine transform, wavelet transform and a direct method. Classification performances, training times and the numbers of nodes of Kohonen network, Restricted Coulomb Energy (RCE) network and the hybrid neural network are presented. To increase the classification performance and to decrease the number of nodes, the hybrid neural network is trained by Genetic Algorithms (GAs). Ten types of ECG beats obtained from the MIT-BIH database and from a real-time ECG measurement system are classified with a success of 98% by using the hybrid neural network structure and discrete cosine transform together.     

Blowing Bubbles for Multi-Scale Analysis and Decomposition of Triangle Meshes

Tools for the automatic decomposition of a surface into shape features will facilitate the editing, matching, texturing, morphing, compression and simplification of three-dimensional shapes. Different features, such as flats, limbs, tips, pits and various blending shapes that transition between them, may be characterized in terms of local curvature and other differential properties of the surface or in terms of a global skeletal organization of the volume it encloses. Unfortunately, both solutions are extremely sensitive to small perturbations in surface smoothness and to quantization effects when they operate on triangulated surfaces. Thus, we propose a multi-resolution approach, which not only estimates the curvature of a vertex over neighborhoods of variable size, but also takes into account the topology of the surface in that neighborhood. Our approach is based on blowing a spherical bubble at each vertex and studying how the intersection of that bubble with the surface evolves. We describe an efficient approach for computing these characteristics for a sampled set of bubble radii and for using them to identify features, based on easily formulated filters, that may capture the needs of a particular application.     

Incorporating Ancestors' Influence in Genetic Algorithms

A new criterion of fitness evaluation for Genetic Algorithms is introduced where the fitness value of an individual is determined by considering its own fitness as well as those of its ancestors. Some guidelines for selecting the weighting coefficients for quantifying the importance to be given to the fitness of the individual and its ancestors are provided. This is done both heuristically and automatically under fixed and adaptive frameworks. The Schema Theorem corresponding to the proposed concept is derived. The effectiveness of this new methodology is demonstrated extensively on the problems of optimizing complex functions including a noisy one and selecting optimal neural network parameters.     

基于几何活动轮廓模型的人脸轮廓提取方法

针对在结构性噪声较严重的情况下 ,常规几何活动轮廓模型无法获得理想分割效果的问题 ,提出一种基于几何活动轮廓模型的人脸轮廓提取方法 ,该方法首先将人脸形状的椭圆性约束作为算子嵌入到几何活动轮廓模型中 ,并利用几何活动轮廓模型提取任意轮廓的优势来快速抽取出图象中类似椭圆的目标边缘 ;然后根据图象中人脸的先验知识 ,通过对检测到的椭圆目标进行进一步验证来找出最终人脸轮廓 .由于采用变分水平集方法做数值计算 ,因此该方法不仅能够自然地处理曲线的拓扑变化和能较精确地提取出图象中的人脸轮廓 ,而且同时可以给出人脸水平旋转的大致角度等信息 .实验结果表明 ,该方法是有效的 .     

一种基于局部结构信息的指纹伪特征滤除算法

自动指纹识别系统中，灰度指纹图象经过预处理过程得到细化二值图象，其中往往含有大量的伪特征，这将对后续的分类、匹配等操作造成不良影响，导致系统识别率下降，为此首先提出一种新的快速纹线跟踪算法——8邻域编码纹线跟踪算法，然后提出一种基于局部结构信息的指纹伪特征滤除算法，该伪特征滤除算法是在纹线跟踪的基础上，提取指纹特征点的若干属性，并结合特征点的局部结构信息，对各种伪特征结构进行识别和滤除，实验结果表明，本方法可以快速、准确、彻底地滤除这些伪特征结构，效果令人满意。     

支持向量机在车牌字符识别中的应用

支持向量机(SVM)是由Vapnik等人提出的一类新型机器学习方法。该文在字符特征提取基础上，应用SVM算法对车牌中的英文字符进行识别，克服了一般的SVM算法识别数字位图时缺乏对相邻空间像素相关性考虑的不足，在满足实时性的条件下获得高识别率。通过与基于字符特征的BP网络识别方案相比较表明，该方案性能远优于神经网络的性能，可很好地解决神经网络方法中无法避免的局部极值问题。实验讨论了在应用SVM算法对字符进行识别时，核函数K和惩罚因子C的选择对识别率的影响问题。