一种识别手写汉字的多分类器集成方法

根据多信源信息处理与字符识别的经验知识,提出了一个识别手写汉字的多分类器 线性集成模型.这个模型不仅考虑到不同的分类器对不同字符识别能力的不同,而且还考虑 了不同的分类器得出的输入字符与参考模板之间相似度的实际大小对判决的影响,及不同分 类器提供的候选字符对判决的支持作用,更重要的是提供了一种通过监督学习,利用计算机 程序自动计算模型参数的方法,因而实现了一个较好的集成系统.同时,本文还提供了三个用 于集成的分类器,它们集成的结果充分显示了本方法的有效性.     

基于Bagging集成学习的字符识别方法

针对字符识别对象的多样性,提出了一种基于Bagging集成的字符识别模型,解决了识别模型对部分字符识别的偏好现象。采用Bagging采样策略形成不同的数据子集,在此基础上用决策树算法训练形成多个基分类器,用多数投票机制对基分类器预测结果集成输出。理论分析与仿真实验结果表明,所提模型相比其他分类方法具有更好的分类能力。     

基于分类器判决可靠度估计的最优线性集成方法

多分类器集成的目标是充分利用每一个分类器的长处，既要发挥它们的最佳性能，又能克服单个分类器的弱点，以达到最优的系统识别性能，该文提出一种基于发类器判决可靠度估计的多分类器集成方法，利用各分类器对当前输入样本的判决兵验概率，实时估计它们的分类判决可靠度，并指导集成权重系数的分配，该方法能够使集成权重更灵活地适应不同的输入变化，充分利用每个分类器当前的识别能力，从而获得更好的集成效果，该文结合最小均方误差准则（MSE）下的最优线性集成方法给出了权重模型和训练算法，并与其它的线性集成模型作了比较，实验表明，不论在最优化准则下的最终误差还是在实际识别性能上，作者提出的方法都取得了最好的效果。     

基于置信度的多分类器互补集成手写数字识别

离线手写数字识别是光学字符识别的一个重要分支,在银行票据识别、邮政编码识别等领域有着广泛的应用。由于单一分类器在识别率上很难达到要求,人们提出了各种集成分类器识别方案。通过对离线手写数字的特征提取,从特征互补的角度出发,采用了最小距离分类器、树分类器和BP网络分类器进行多分类器互补集成,提出了基于置信度的多分类器互补集成方法。通过实验对比,基于置信度的多分类器互补集成手写数字识别在识别率和识别速度上达到了满意的结果。     

基于笔划特征的单字符汉字字体识别

在文档电子化的文本自动分析、理解和识别过程中,除了有关文档内容的字符识别外,还必须解决字体识别问题.字体识别不仅是版面分析、理解和恢复的重要依据,还有助于实现高性能字符识别系统.有别于目前基于多个字符组成的文本块的字体识别方法,本文提出了一种基于单个汉字字符的字体识别方法.在单个汉字字符上提取两类特征:笔划属性特征和笔划分布特征,分别构成两个分类器对单个汉字字符进行字体识别,并集成两个分类器的结果得到最终的识别结果.我们使用的笔划属性特征分类器是文本无关的,而笔划分布特征分类器是文本相关的,集成的分类器属于文本相关的字体识别分类器.我们在包含7种字体的样本集上进行了测试,测试结果显示基于单字的字体识别率达到94.48%.     

竞争监督学习法在集成型识别系统中的应用

在集成方法中,神经网络集成方法对研制集成型模式识别系统是有效的.但是,单个 子分类器和集成网络的性能对集成系统的整体识别效果都有影响.因此,要进一步提高系统 性能也必须改进子分类器和集成网络.文中采用竞争监督学习法,构造一个网络集成系统,用 于手写数字字符识别.实验证明,该方法的确能够改进系统的收敛速度和泛化能力.     

一种用于手写体汉字识别的候选字加权多分类器集成方法

提出了一种基于候选字加权的多分类器集成方法,并将其应用于手写体汉字的识别研究中。利用４种不同的特征提取方法构造了４个独立的分类器;利用不同分类器各候选字加权处理得到的置信度函数来构造集成函数,从而将４个独立的分类器集成为一个多分类器系统。通过实验分析了几种分类器集成的方法,验证了具有一定互补性的多分类器集成对手写体汉字的识别率有较大的提高。实验结果表明、所提出的方法是行之有效的。     

离线签名验证中的多分类器集成

在离线签名验证的分类器设计中,为了减少特征向量分布不均和维数过高对实验结果的影响,给出一种多分类器集成的方法.根据特征向量数量级的不同进行分组,各组分类器自适应地确定分类器权重,通过投票表决得出集成判决结果.实验结果表明,通过分组和加权后,分类正确率有明显提高.     

基于主分量分析的手写数字字符识别

针对手写数字字符识别中由于书写习惯和风格的不同，造成字符模式不稳定的问题，提出了一种图像预处理方法．首先采用数学形态学通过细化和膨胀，统一字符笔画的粗细，并使字符的局部特征得到改善；然后利用主分量分析法(PCA)抽取字符特征，估计字符的重建模型，并通过对重建模型的误差分析进行字符识别；最后通过对美国国家邮政局USPS字库中全部数字字符完整的识别实验，证实了算法的鲁棒性和准确性．     

AdaBoost算法在车牌字符识别中的应用

提出了一种基于AdaBoost的车牌字符自动识别算法。AdaBoost是一种构建准确分类器的学习算法，它将一族弱学习算法通过一定规则结合成为一个强学习算法，从而通过样本训练得到一个识别准确率理想的分类器，将之用于车牌字符识别，对车牌图像进行实验，对车牌字符样本进行特征提取，用特征来训练有效分类器，用MATLAB完成了对车牌照数字识别的模拟，结果证实此算法对车牌字符识别有一定准确性，具有良好的效果。     

Software quality prediction using fuzzy integration: a case study

Given the complexity of many contemporary software systems, it is often difficult to gauge the overall quality of their underlying software components. A potential technique to automatically evaluate such qualitative attributes is to use software metrics as quantitative predictors. In this case study, an aggregation technique based on fuzzy integration is presented that combines the predicted qualitative assessments from multiple classifiers. Multiple linear classifiers are presented with randomly selected subsets of automatically generated software metrics describing components from a sophisticated biomedical data analysis system. The external reference test is a software developer’s thorough assessment of complexity, maintainability, and usability, which is used to assign corresponding quality class labels to each system component. The aggregated qualitative predictions using fuzzy integration are shown to be superior to the predictions from the respective best single classifiers.     

Novel mislabeled training data detection algorithm

As a kind of noise, mislabeled training data exist in many applications. Because of their negative effects on learning, many filter techniques have been proposed to identify and eliminate them. Ensemble learning-based filter (EnFilter) is the most widely used filter which employs ensemble classifiers. In EnFilter, first the noisy training dataset is divided into several subsets. Each noisy subset is then checked by the multiple classifiers which are trained based on other noisy subsets. It is noted that since the training data used to train multiple classifiers are noisy, the quality of these classifiers cannot be guaranteed, which might generate poor noise identification result. This problem is more serious when the noise ratio in the training dataset is high. To solve this problem, a straightforward but effective approach is proposed in this work. Instead of using noisy data to train the classifiers, nearly noise-free (NNF) data are used since they are supposed to train more reliable classifiers. To this end, a novel NNF data extraction approach is also proposed. Experimental results on a set of benchmark datasets illustrate the utility of our proposed approach.

     

A class decomposition approach for GA-based classifiers

Genetic algorithm (GA) has been used as a conventional method for classifiers to evolve solutions adaptively for classification problems. In this paper, a new approach using class decomposition is proposed to improve the performance of GA-based classifiers. A classification problem is fully partitioned into several class modules in the output domain and each module is responsible for solving a fraction of the original problem. These modules are trained in parallel and independently and the results obtained are integrated and evolved further for a final solution. A scheme based on Fisher's linear discriminant (FLD) computation is used to estimate the difficulty of separating two classes. Based on the FLD information derived, different integration approaches are proposed and their performance is compared. The experiment results on a benchmark data set show that class decomposition can achieve higher classification rate than the normal GA and FLD-based integration improves the classification accuracy further.     

Estimating 3D hand pose using hierarchical multi-label classification

This paper presents an analysis of the design of classifiers for use in a hierarchical object recognition approach. In this approach, a cascade of classifiers is arranged in a tree in order to recognize multiple object classes. We are interested in the problem of recognizing multiple patterns as it is closely related to the problem of locating an articulated object. Each different pattern class corresponds to the hand in a different pose, or set of poses. For this problem obtaining labelled training data of the hand in a given pose can be problematic. Given a parametric 3D model, generating training data in the form of example images is cheap, and we demonstrate that it can be used to design classifiers almost as good as those trained using non-synthetic data. We compare a variety of different template-based classifiers and discuss their merits.     

Adaptive classifier integration for robust pattern recognition

The integration of multiple classifiers promises higher classification accuracy and robustness than can be obtained with a single classifier. This paper proposes a new adaptive technique for classifier integration based on a linear combination model. The proposed technique is shown to exhibit robustness to a mismatch between test and training conditions. It often outperforms the most accurate of the fused information sources. A comparison between adaptive linear combination and non-adaptive Bayesian fusion shows that, under mismatched test and training conditions, the former is superior to the latter in terms of identification accuracy and insensitivity to information source distortion.     

基于专家域的多层分类器融合

论文提出了一种基于专家域的多层分类器融合模型,专家指不同专长之单分类器。模型思想来自医院诊断流程,模型首先训练n个专家,之后将样本空间按专家专长划分专家域。对于待测样本,先将样本指派到合适的专家域,然后再由指定的专家对样本进行分类。用这种算法对UCI的标准数据集进行分类,实验结果显示,该算法得到比其他算法更低的分类误差,显著提高了分类器的性能。     

多分类器选择集成方法

针对目前人们对分类性能的高要求和多分类器集成实现的复杂性,从基分类器准确率和基分类器间差异性两方面出发,提出了一种新的多分类器选择集成算法。该算法首先从生成的基分类器中选择出分类准确率较高的,然后利用分类器差异性度量来选择差异性大的高性能基分类器,在分类器集成之前先对分类器集进行选择获得新的分类器集。在UCI数据库上的实验结果证明,该方法优于bagging方法,取得了很好的分类识别效果。     

A mixed integer optimisation model for data classification

In this work, a mixed integer linear programming (MILP) model is proposed for the multi-class data classification problem using a hyper-box representation. The latter representation is particularly suitable for capturing disjoint data regions. The objective function used is the minimisation of the total number of misclassified data samples. In order to improve the training and testing accuracy of our approach, an iterative solution procedure is developed to assign potential multiple boxes to each single class. Finally, the applicability of the proposed approach is demonstrated through a number of illustrative examples. According to the computational results obtained, the proposed optimisation-based approach is competitive in terms of prediction accuracy when compared with various standard classifiers.     

Knowledge integration in a multiple classifier system

This paper introduces a knowledge integration framework based on Dempster-Shafer's mathematical theory of evidence for integrating classification results derived from multiple classifiers. This framework enables us to understand in which situations the classifiers give uncertain responses, to interpret classification evidence, and allows the classifiers to compensate for their individual deficiencies. Under this framework, we developed algorithms to model classification evidence and combine classification evidence form difference classifiers, we derived inference rules from evidential intervals for reasoning about classification results. The algorithms have been implemented and tested. Implementation issues, performance analysis and experimental results are presented.