手写字符识别的非线性主动判别函数

提出了用于手写字符识别的非线性主动判别函数,是线性主动判别函数在手写字符非线性变化情况下的推广。该方法利用Kernel PCA分析捕捉和表示这种非线性变化。将输入空间非线性映射为特征空间,在特征空间的主子空间中生成最优主动原型模板,其与字符特征向量在特征空间主子空间的投影之间的距离即为非线性主动判别函数;同时,基于最小分类错误准则对该函数进行了优化。实验结果表明,非线性主动判别函数获得了比线性主动判别函数更高的识别率。     

基于轮廓和统计特征的手写体数字识别

提出了数字规范化模板特征,并利用这一特征与轮廓分段特征相结合对手写体数字进行识别。首先使用基于轮廓分段特征的分类器进行识别,通过提高拒识率获得高可靠性的分类结果。然后由基于数字规范化模板特征的分类器对前一级分类器的拒识样本分类。实验结果表明分别基于这两个特征的分类器在分类结果上具有较强的互补性。实验的数据为真实支票上采集的10000个手写体数字样本,该方法的识别率为98.06%。     

一种基于AP的仿生模式识别方法

提出了一种基于仿射传播聚类(Affinity Propagation Clustering,AP Clustering)和仿生模式识别理论(Biomimetic Pattern Rccognition,BPR)的识别方法。该方法通过AP聚类选择代表训练样本,依据仿生模式识别理论构建并划分样本空间,通过计算待识样本到各特征子空间的相对距离,根据其所处空间进行分类识别。在因空间重叠造成拒识的情况下,通过计算基于类条件的后验概率对样本进行相对区别。在Concordia大学CENPARMI手写体数字库与南京理工大学手写金额库上进行了实验,结果表明,该方法在识别率方面优于传统的分类器。     

基于模板匹配的手写数字决策分类器设计

针对手写数字识别提出一种基于模板匹配决策分类器设计方法。就该方法下的模式识别分类器设计进行详细论述,给出该分类器算法实现。该算法在对手写的数字图像进行预处理的基础上从待识剐的手写数字图像中提取若干特征量与事先建立的标准模板库中模板对应的特征量进行比较,计算待识别图像和标准模板特征量之间的距离,用最小距离法判定其所属类。实验结果表明,该决策分类器算法实现容易,匹配速度快,保证字符识别的正确率。     

一种基于数学形态学的手写汉字方向特征提取方法

提出了一种新的基于数学形态学的手写体汉字方向特征提取的方法。该方法首先提出了一种利用数学形态学对手写汉字做笔划宽度归一化的方法，然后应用数学形态学的方向模板提取汉字在横、竖、撇、捺4个方向上的模式图像，最后，将弹性网格作用在4幅方向模式图像中，统计每个网格中的黑像素分布密度，得到汉字的方向特征。使用简单的距离分类器对120套3775个汉字手写体汉字进行识别实验，识别率达到8547％。     

一种基于独立分量分析的钞票识别算法

提出一种基于独立分量分析的钞票识别算法.先对钞票图像作感兴趣区域(ROI)切割;接着对ROI图像作标准化、白化预处理;然后采用基于负熵独立性判据的固定点方法(FastICA)对预处理后的ROI图像做ICA分离,提取独立基图像,进而获得钞票的特征空间,并构建特征模板;通过计算待识别目标与特征模板的距离实现识别.以第五套人民币作为实验对象进行实验,实验结果表明方法的有效性.     

纹理特征和两级分类器相结合的人脸识别

为了提高人脸识别率和识别效率,提出一种纹理特征和两级分类器相结合的人脸识别方法。采用灰度共生矩阵表示人脸图像的纹理特征,计算待识别人脸图像与模板间欧式距离,采用拒识阈值进行评判,如果人脸图像归属类别清楚,则采用欧式距离分类器进行识别,否则将待识人脸图像送入SVM分类器进行识别,采用ORL人脸数据库和Yale人脸数据库进行仿真实验。仿真结果表明,相对于单一人脸识别器,两级分类器不仅提高了人脸识别效率,而且提高了人脸识别率,具有更好的人脸识别性能。     

一种基于主成分分析的钞票识别算法

提出一种基于PCA的钞票识别算法。通过对经预处理后的钞票图像作感兴趣区域(ROI)切割,获得ROI图像;然后对ROI图像做K-L变换提取钞票图像的主成分,构造钞票的特征空间;利用训练样本在特征空间中的投影向量构建特征模板。通过计算待识别目标与特征模板的最小距离来完成识别。以第五套人民币作为实验对象进行实验,实验结果表明本文提出方法的有效性。     

基于BP神经网络的手写体数字识别

手写体数字识别是多年来的研究热点,也是字符识别中的一个特别问题.由于手写体数字字体变化很大,传统的识别方法很难达到高的识别率.针对传统的数字识别方法的复杂性和局限性,提出了一种基于BP神经网络的手写体数字的识别方法.该方法在提取手写体数字点特征、笔划密度特征基础上,利用改进的BP神经网络进行训练识别.经实验,识别率达94%.实验结果表明,该方法对手写体数字识别效果良好,不仅简化了传统识别的繁杂性,而且提高了识别的准确性.     

英汉翻译中短语的识别与分析

本文在讨论了英文事的形态结构特征的基础上,提出了一种以短语模式空间匹配为基础的短语识别算法。该算法通过对短语的原型描述与输入文本中的全部可能路径进行递增模式匹配来识别具有外部形态约束和框架结构的复杂短语构。本文最后讨论了短语模板和基于复杂特征的短语描述方式。     

Improved support vector classification using PCA and ICA feature space modification

An approach that unifies subspace feature selection and optimal classification is presented. Independent component analysis (ICA) and principal component analysis (PCA) provide a maximally variant or statistically independent basis for pattern recognition. A support vector classifier (SVC) provides information about the significance of each feature vector. The feature vectors and the principal and independent component bases are modified to obtain classification results which provide lower classification error and better generalization than can be obtained by the SVC on the raw data and its PCA or ICA subspace representation. The performance of the approach is demonstrated with artificial data sets and an example of face recognition from an image database.     

面向连续叠写的高精简中文手写识别方法研究

连续手写识别是中文手写输入技术的核心,自然、快捷地输入中文信息一直是模式识别乃至人工智能领域追求的目标。提出了一种有效克服小屏幕限制的连续叠写汉字识别方法。该方法基于切分-识别集成的解码框架,先使用过切分算法处理输入的书写轨迹;然后启用一种新颖的感知机算法判定字符的边界;随后采用来自字符分类模型、几何模型和语言模型的多种上下文信息进行路径解码。为适应不同类型的移动终端,特别提出了一种高效压缩字符分类模型的方法,以有效减少字符识别过程对存储和内存的占用。该识别方法已在Android平台上部署,并进行了大规模的测试实验。实验结果证实了该识别方法的性能和效率。     

Feature interaction in subspace clustering using the Choquet integral

Subspace clustering has recently emerged as a popular approach to removing irrelevant and redundant features during the clustering process. However, most subspace clustering methods do not consider the interaction between the features. This unawareness limits the analysis performance in many pattern recognition problems. In this paper, we propose a novel subspace clustering technique by introducing the feature interaction using the concepts of fuzzy measures and the Choquet integral. This new framework of subspace clustering can provide optimal subsets of interacted features chosen for each cluster, and hence can improve clustering-based pattern recognition tasks. Various experimental results illustrate the effective performance of the proposed method.     

手写汉字识别的非线性动态部件模板

对以前提出的非线性动态手写模板加以改进并用于手写汉字的部件识别.在训练阶 段,核-主元分析用来捕捉非线性的手写变化.于是,只需改变少量的形状参数就可获得动态变 形的模板.在识别阶段,遗传算法取代了原始的动态通道算法去寻找最优的形状参数.我们对覆 盖2154个汉字类别的200个部件进行了实验,对不用人书写的430,800个测试样本的部件识 别率达97.4%.与现有的代表性部件方法比较也显示本文的方法效果最好.     

A Hilbert warping method for handwriting gesture recognition

We propose a novel sequence alignment algorithm for recognizing handwriting gestures by a camera. In the proposed method, an input image sequence is aligned to the reference sequences by phase-synchronization of analytic signals which are transformed from original feature values. A cumulative distance is calculated simultaneously with the alignment process, and then used for the classification. A major benefit of this method is that over-fitting to sequences of incorrect categories is restricted. The proposed method exhibited higher recognition accuracy in handwriting gesture recognition, compared with the conventional dynamic time warping method which explores optimal alignment results for all categories.     

Locality preserving embedding for face and handwriting digital recognition

Most supervised manifold learning-based methods preserve the original neighbor relationships to pursue the discriminating power. Thus, structure information of the data distributions might be neglected and destroyed in low-dimensional space in a certain sense. In this paper, a novel supervised method, called locality preserving embedding (LPE), is proposed to feature extraction and dimensionality reduction. LPE can give a low-dimensional embedding for discriminative multi-class sub-manifolds and preserves principal structure information of the local sub-manifolds. In LPE framework, supervised and unsupervised ideas are combined together to learn the optimal discriminant projections. On the one hand, the class information is taken into account to characterize the compactness of local sub-manifolds and the separability of different sub-manifolds. On the other hand, at the same time, all the samples in the local neighborhood are used to characterize the original data distributions and preserve the structure in low-dimensional subspace. The most significant difference from existing methods is that LPE takes the distribution directions of local neighbor data into account and preserves them in low-dimensional subspace instead of only preserving the each local sub-manifold’s original neighbor relationships. Therefore, LPE optimally preserves both the local sub-manifold’s original neighborhood relationships and the distribution direction of local neighbor data to separate different sub-manifolds as far as possible. The criterion, similar to the classical Fisher criterion, is a Rayleigh quotient in form, and the optimal linear projections are obtained by solving a generalized Eigen equation. Furthermore, the framework can be directly used in semi-supervised learning, and the semi-supervised LPE and semi-supervised kernel LPE are given. The proposed LPE is applied to face recognition (on the ORL and Yale face databases) and handwriting digital recognition (on the USPS database). The experimental results show that LPE consistently outperforms classical linear methods, e.g., principal component analysis and linear discriminant analysis, and the recent manifold learning-based methods, e.g., marginal Fisher analysis and constrained maximum variance mapping.     

基于书写摩擦力的签名识别方法

由于动态测量坐标系的存在, 使得数字笔在不同的握笔姿态下, 引起书写平面内摩擦力的分解不同. 由此本文提出采用书写摩擦力进行在线签名识别. 该方法将摩擦力的方向在直角坐标系内进行特征量表示, 并与三维书写力的大小相结合, 实现概率值和距离测度信度级的融合. 在进行真假签名的判别中, 加入快速剔除的方法, 进行两级判别, 以提高识别效率. 实验结果表明, 基于签名书写力大小和方向特征的融合识别方法互相补充, 相对于单个方法具有更高的识别率, 系统性能得到显著改善.