基于主曲线的脱机手写数字结构特征分析及选取

要提高脱机手写数字识别的识别率,关键是特征的提取与选择．主曲线是主成分分析的非线性推广,它是通过数据分布“中间”并满足“自相合”的光滑曲线．它较好地反映了数据分布的结构特征．在数字特征选取中,首先将主曲线用于训练数据的特征提取;其次在详细分析数字主曲线的结构特点的基础上,选择出用于数字识别的粗分类、细分类特征;最后在对手写数字进行识别时,先进行粗分类再进行细分类．所提方法在Concordia大学的CENPARMI手写体数字数据库上的实验结果表明：利用这些特征能有效区分相似字符,提高了手写数字的识别率,为脱机手写数字识别的研究提供了一条新途径。     

基于主曲线算法的手写字符特征分析与提取

模式特征的提取与选择是提高手写体字符识别率的关键因素。主曲线是主成分分析的非线性推广,它是通过数据分布“中间”并满足“自相合”的光滑曲线,能够很好地描述数据分布的结构特征。利用软K段主曲线算法提取训练数据的特征,在分析手写体字符结构特点的基础上,选出手写体字符识别所使用的粗分类与细分类特征,利用这些分类特征对手写字符进行识别。该方法在CEDAR手写体数字和字符数据库上的实验表明：选取的分类特征能够有效区分相似的手写体字符,提高手写字符的识别率,为脱机手写字符识别研究提供了一种新的方法。     

基于主曲线的相似字符模糊分类方法

要提高手写字符的识别率,抽取方法,它是通过数据分布"中间"并满足"自相合"的光滑曲线,较好地反映了数据分布的结构特征.本文尝试用主曲线这种新的方法来提取手写字符的结构特征,并基于这些特征来对相似字符进行模糊分类.所提方法在CE-DAR和OCRD手写体字符数据库上的实验结果表明:该方法不但是可行的,而且能有效提高相似字符的识别率.它为字符识别的研究提供了一条新途径.     

基于软K段主曲线算法的字符特征提取研究及实现

要提高脱机手写字符识别的识别率,关键是特征的提取。主曲线是主成分分析的非线性推广,是通过数据分布“中间”并满足“自相合”的光滑曲线。通过对现有主曲线算法分析可知：软K段主曲线算法对提取出分布在弯曲度很大或相交曲线周围的数据的主曲线效果较好。因此本文尝试用谊主曲线算法来提取脱机手写字符的结构特征。实验结果表明,利用该主曲线算法来提取脱机手写字符的结构特征不但是可行的,而且取得较好的实验效果。它为脱机手写字符特征提取的研究提供了一条新途径。     

基于层次分类的脱机手写字符识别

人类在进行字符识别活动时,会根据对象复杂度的不同,采用不同的识别方法.对于结构简单的字符,利用宏观整体信息识别;对于易混淆的形近字,利用微观具体信息区分.为了模拟人类智能进行字符识别活动的过程,设计了一种基于层次分类的脱机手写字符识别算法.该算法将分类器划分为宏观层和微观层,宏观层模拟简单字符识别过程,利用基于梯度的统计特征描述整体信息,完成识别;微观层模拟形近字识别过程,利用基于主曲线的结构特征描述具体信息,完成区分.算法还引入了可信度概念,用以量度推理过程及识别结果的不确定性程度.给出了形近字的定义及区分规则.实验表明,提出的算法有效地提高了脱机手写字符的识别率,对形近字的区分效果尤佳.     

基于多特征集成分类器的脱机满文识别方法

为提高脱机满文手写字体的识别率,提出了基于BP网络的多特征集成分类器识别方法.对扫描成图像的手写满文进行预处理,切分出满文字元;分别提取满文字元的投影特征、链码特征以及端点和交叉点特征,并对这三类特征及其相互组合进行分类识别;通过隐马尔科夫算法对识别结果进行后处理,进一步提高识别的精度.实验结果表明,集成分类器的识别率要比单个特征的识别率要高,同时集成分类器中的特征类别越多,识别效果越好.     

基于多分类器集成的维吾尔文联机手写字母识别

多分类器组合能够在一定程度上弥补单个分类器的缺陷,因此它在模式识别中得到了广泛应用。深入调研国内外联机手写识别技术的研究动态,结合维吾尔文字母的独特书写风格,研究了基于多分类器集成的维吾尔语联机手写字母识别。利用5种不同的特征提取方法构造了5个独立的维吾尔语字母分类识别器,采用了等权投票和不等权投票等两种策略将5种维吾尔语字母分类识别器进行了有效组合。其中,单分类器采用了基于动态时间弯折（DTW）匹配距离的最近邻分类方法。实验结果表明,提出的集成策略的识别率明显高于单分类器的识别率,而且为特征的综合集成提供了多种有效途径。     

自动评阅系统中手写英文字母识别算法研究

该文就自动评阅系统中手写英文字母的识别问题提出一种简单高效的识别方法。对预处理之后的图像进行水平和垂直投影,通过提取相关的特征信息对目标字母进行分类,经过几次分类完成对字母的识别。实验证明该方法可以较快的完成识别,并具有较高的准确性。     

SVM多值分类器在脱机手写体相似汉字识别中的应用

相似字的普遍存在是影响脱机手写体汉字识别率低的主要原因之一。论文研究了支持向量机(SVM)多值分类器在手写相似汉字识别中的应用,所提出的方法采用了小波弹性网格技术提取汉字的特征,通过实验比较了三种不同的SVM分类器组合策略的分类效果。     

基于双射变换的脱机手写数字识别方法

针对传统特征提取和分类方法速度慢、稳定性差、识别率低等特点,提出了一种基于外围结构特征提取的手写数字识别方法。该方法多次少量地提取经过双射变换后的图像外围结构特征,对每一次提取的特征结合BP神经网络生成相应的分类器,对不同特征的分类结果进行融合得出手写数字的识别结果。实验结果表明,该特征提取方法实现简单,运算量小,大大提高了脱机手写数字的识别率和效率。     

基于主曲线的指纹细节特征提取方法

要提高指纹识别的识别率,关键是指纹特征的提取。主曲线是主成份分析的非线性推广,它是通过数据分布“中间’’并满足“自相合”的光滑曲线,较好地反映了数据分布的结构特征。本文尝试使用主曲线这种新的方法来提取指纹的细节特征。实验结果表明利用主曲线来提取指纹的结构特征是可行的,它为指纹特征提取的研究提供了一条新途径。     

HS主曲线的数学特性

主曲线被定义作穿过多维数据分布“中间”的满足“自相合”的光滑曲线,它是第一主成分的非线性推广,第一主成分是对数据集的一维线性最优描递。HS主曲线强调非参数模型,对其参数无关性本文给出了具体证明。同时为了全面理解主曲线,本文以空间主曲线为例,分析了它的横截性质。     

主曲线算法初始值选取的分析

主曲线是一种用于数据压缩和特征提取的有效方法,是对主成分分析的非线性推广。由于主曲线与主成分的密切联系,主曲线生成算法通常以第二主成分线做初始值。然而实验发现第一主成分未必是算法初始化的最佳选择。本文将以HS算法和多边形算法为例,就初始值的选取对生成主曲线的影响做出分析并通过实验得出结论：HS算法以原点作初值效果较好,多边彤算法应根据数据点集的不同结构选择合适的初值。     

主曲线研究综述

主曲线（principal curves)是第一主成分的非线性推广，第一主成分是对数据集的一维线性最优描述，主轴线强调寻找通过数据分布的“中间（middle)并满足“自耦合”的光滑一维曲线，其理论基础是寻找嵌入高维空间的非欧氏低维流形，该文着重介绍了主曲线发展的动机，理论基础，典型的主曲线方法和算法实现及其不同领域的应用，并对存在的问题进行了分析。     

基于主曲线的微阵列数据分类

提出了一种基于主曲线(principal curves)的微阵列数据分类方法(PC)。主曲线是第一主成分的非线性推广,它是数据集合的“骨架”,数据集合是主曲线的“云”。基于主曲线的微阵列数据分类方法,首先利用专门设计的算法在训练数据集上计算出每类样本的主曲线,然后根据测试样本与各类样本主曲线距离的期望方差来确定测试样本所属的类别。实验结果表明,该分类方法在进行小样本微阵列数据分类时性能优于现有的方法。     

Classification of brain signal (EEG) induced by shape-analogous letter perception

Visual perception of English letters involves different underlying brain processes including brain activity alteration in multiple frequency bands. However, shape analogous letters elicit brain activities which are not obviously distinct and it is therefore difficult to differentiate those activities. In order to address discriminative feasibility and classification performance of the perception of shape-analogous letters, we performed an experiment in where EEG signals were obtained from 20 subjects while they were perceiving shape analogous letters (i.e., ‘p’, ‘q’, ‘b’, and ‘d’). Spectral power densities from five typical frequency bands (i.e., delta, theta, alpha, beta and gamma) were extracted as features, which were then classified by either individual widely-used classifiers, namely k-Nearest Neighbors (kNN), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), Random Forest (RF) and AdaBoost (ADA), or an ensemble of some of them. The F-score was employed to select most discriminative features so that the dimension of features was reduced. The results showed that the RF achieved the highest accuracy of 74.1% in the case of multi-class classification. In the case of binary classification, the best performance (Accuracy 86.39%) was achieved by the RF classifier in terms of average accuracy across all possible pairs of the letters. In addition, we employed decision fusion strategy to exert complementary strengths of different classifiers. The results demonstrated that the performance was elevated from 74.10% to 76.63% for the multi-class classification and from 86.39% to 88.08% for the binary class classification.     

Integration of multitemporal/polarization C‐band SAR data sets for land‐cover classification

This paper investigates the potential of multitemporal/polarization C‐band SAR data for land‐cover classification. Multitemporal Radarsat‐1 data with HH polarization and ENVISAT ASAR data with VV polarization acquired in the Yedang plain, Korea are used for the classification of typical five land‐cover classes in an agricultural area. The presented methodologies consist of two analytical stages: one for feature extraction and the other for classification based on the combination of features. Both a traditional SAR signal property analysis‐based approach and principal‐component analysis (PCA) are applied in the feature extraction stage. Special concerns are in the interpretation of each principal component by using principal‐component loading. The tau model applied as a decision‐level fusion methodology can provide a formal framework in which the posteriori probabilities derived from different sensor data can be combined. From the case study results, the combination of PCA‐based features showed improved classification accuracy for both Radarsat‐1 and ENVISAT ASAR data, as compared with the traditional SAR signal property analysis‐based approach. The integration of PCA‐based features based on multiple polarization (i.e. HH from Radarsat‐1, and both VV and VH from ENVISAT ASAR) and different incidence angles contributed to a significant improvement of discrimination capability for dry fields which could not be properly classified by using only Radarsat‐1 or ENVISAT ASAR data, and thus showed the best classification accuracy. The results of this case study indicate that the use of multiple polarization SAR data with a proper feature extraction stage would improve classification accuracy in multitemporal SAR data classification, although further consideration should be given to the polarization and incidence angle dependency of complex land‐cover classes through more experiments.     

基于随机森林的高光谱遥感图像分类

为了充分利用高光谱图像的光谱信息和空间结构信息,提出了一种新的基于随机森林的高光谱遥感图像分类方法,首先,利用主成分分析降低数据的维数,并对主成分进行独立成分分析提取其光谱特征,同时消除像元的空间相关性,再采用形态学分析提取像元的空间结构特征,然后,根据像元的谱域和空域特征分别构造随机森林,并引入空间连续性对像元点的预测结果进行约束修正,最后由投票机制决定最后的分类结果。在AVIRIS和ROSIS高光谱图像上的实验结果表明,所提方法的分类性能要优于传统的高光谱图像分类方法,且分类精度高于基于单一特征的方法。     

Piecewise linear skeletonization using principal curves

Proposes an algorithm to find piecewise linear skeletons of handwritten characters by using principal curves. The development of the method was inspired by the apparent similarity between the definition of principal curves (smooth curves which pass through the "middle" of a cloud of points) and medial axes (smooth curves that run equidistantly from the contours of a character image). The central fitting-and-smoothing step of the algorithm is an extension of the polygonal line algorithm, which approximates principal curves of data sets by piecewise linear curves. The polygonal line algorithm is extended to find principal graphs and complemented with two steps specific to the task of skeletonization: an initialization method to capture the approximate topology of the character, and a collection of restructuring operations to improve the structural quality of the skeleton produced by the initialization method. An advantage of our approach over existing methods is that we optimize the skeleton graph by minimizing an intuitive and explicit objective function that captures the two competing criteria of smoothing the skeleton and fitting it closely to the pixels of the character image. We tested the algorithm on isolated handwritten digits and images of continuous handwriting. The results indicated that the proposed algorithm can find a smooth medial axis in the great majority of a wide variety of character templates and that it substantially improves the pixel-wise skeleton obtained by traditional thinning methods