基于多级纹理频谱特征与PCA的人脸识别算法

针对主成分分析(PCA)算法在人脸识别中识别率低的问题,提出一种图像纹理频谱特征与PCA相结合的人脸识别算法。该算法利用纹理单元算子提取人脸图像纹理频谱特征,然后用PCA对所提取的特征降维,最后利用最近邻(KNN)分类器进行人脸识别。在ORL人脸库和Yale人脸库上对所提出的算法进行了测试,识别率均高于PCA、模块化二维PCA(M2DPCA)等方法,分别为96.5%和95%。实验结果表明了该算法的有效性和准确性。     

加权小波和流形正则化的NMF融合的人脸识别

为了获取更充分的人脸特征信息以提高识别性能,应用加权小波变换和流形正则化非负矩阵分解的方法实现人脸识别。采用小波变换,提取训练样本人脸图像的加权高频分量和低频分量的特征信息;应用流形正则化非负矩阵分解方法,在保持人脸特征数据原始几何结构和局部特征的基础上获取最终的识别特征;利用最近邻方法进行分类识别。将该算法在ORL人脸库和YALE人脸库上进行测试验证,结果表明,与传统的非负矩阵分解方法相比,其识别率高出5%左右,且计算时间很低,说明该方法耗时短,效率高。     

基于类别平均距离的加权KNN分类算法

本文提出了一种改进的KNN分类算法,利用样本集合中同类别样本点间距离都十分接近的特点辅助KNN算法分类．将待分类样本点的K个最近邻样本点分别求出样本点所属类别的类别平均距离和样本点与待分类样本点距离的差值比,如果大于一个阈值,就将该样本点从K个最近邻的样本点中删除,再用此差值比对不同类别的样本点个数进行加权后执行多数投票,来决定待分类样本点所属的类别．改进后的KNN算法提高了分类的精度,并且时问复杂度与传统KNN算法相当．     

基于正弦变换的人脸姿态校正及识别研究

姿态变化是影响人脸识别率的一个至关重要的因素,也是人脸识别问题中一个待解决的难题。当测试样本具有一定的姿态变化后,识别率会急剧下降。针对此问题,提出了利用正弦变换（Sine Transform,ST）对待识别的姿态图像进行姿态校正,虚拟出对应的正面人脸的方法。使用经典算法进行特征提取、最近邻分类器进行分类识别验证,得到了较好的结果。在FERET人脸库上的实验表明,该方法能够在一定程度上克服姿态变化的影响,平均识别率最高可提高17%。     

基于k-最近邻图的小样本KNN分类算法

提出一种基于k-最近邻图的小样本KNN分类算法。通过划分k-最近邻图,形成多个相似度较高的簇,根据簇内已有标记的数据对象来标识同簇中未标记的数据对象,同时剔除原样本集中的噪声数据,从而扩展样本集,利用该新样本集对类标号未知数据对象进行类别标识。采用标准数据集进行测试,结果表明该算法在小样本情况下能够提高KNN的分类精度,减小最近邻阈值k对分类效果的影响。     

基于对称非迭代双边2DPCA的人脸识别

结合人脸图像的对称性在非迭代双边二维主成分分析（NIB2DPCA）的基础上, 提出了对称非迭代双边二维主成分分析（SNIB2DPCA）的人脸识别方法。该方法引入镜像变换, 根据奇偶分解原理分别生成奇、偶对称样本, 用NIB2DPCA分别对奇偶对称样本提取特征, 通过奇偶加权因子对奇偶对称样本的特征矩阵进行组合得到最终的分类特征矩阵, 最后用最近邻分类器分类。在Yale、ORL和YaleB人脸库上的实验表明该方法不仅显著提高了识别率, 而且对光照影响有一定的鲁棒性。     

基于稀疏表示的低分辨率人脸疲劳表情识别

为了有效提高低分辨率图像的人脸疲劳表情识别性能,提出一种基于稀疏表示的低分辨率人脸疲劳表情的识别方法。首先,采用肯德尔和谐系数可信度分析法构建了低分辨率人脸疲劳表情图像库TIREDFACE。其次,通过图像库中的低分辨率样本疲劳表情图像进行稀疏表示,再利用压缩感知理论寻求低分辨率测试样本的最稀疏解,采用求得的最稀疏解实现低分辨率人脸疲劳表情的分类。在低分辨率人脸视觉特征的疲劳表情图像库TIREDFACE的实验测试结果表明,将该方法用于低分辨人脸疲劳表情识别,性能优于线性法、最近邻法、支持向量机以及最近邻子空间法。可见,该方法用于低分辨率人脸疲劳表情识别时识别效果较好,精确度较高。     

基于主动表观模型的稀疏聚类人脸识别算法

在复杂的非人脸成分干扰以及训练样本过大、训练样本之间相似度较高的条件下,原始稀疏表示分类(SRC)算法识别准确率较低。针对上述问题,提出一种基于主动表观模型的稀疏聚类(CS-AAM)人脸识别算法。首先,利用主动表观模型快速、准确地对人脸特征点进行定位,获取主要人脸信息;然后,对训练样本进行K-means聚类,将相似程度高的图像分为一类,计算聚类中心,将该中心作为原子构造过完备字典并进行稀疏分解;最后,计算稀疏系数和重构残差对人脸图像进行分类、识别。将该算法与最近邻(NN)、支持向量机(SVM)、稀疏表示分类(SRC)、协同表示分类(CRC)人脸识别算法在ORL和Extended Yale B人脸数据库上对不同样本数及不同维数的人脸图像分别进行识别率测试,在相同样本数或相同维数情况下CS-AAM算法识别率均高于其他算法。在ORL人脸库中选取样本数为210时,相同维数条件下CS-AAM算法识别率为95.2%;在Extended Yale B人脸库上选取样本数为600时,CS-AAM算法识别率为96.8%。实验结果表明,该算法能够有效地提高人脸图像的识别准确率。     

KNN-均值算法

基于传统KNN的弱点．提出一种新的改进方法：在训练样本中取与待测样本最近邻的K个样本;把这K个样本按类划分取每类的平均样本;计算每个类的平均样本与待测样本的欧氏距离,并将样本归为距离最小的那一类。在南京理工大学NUST603HW手写体汉字库以及Concordia大学的CENPARMI手写体阿拉伯数字数据库上的试验结果表明,新方法较传统KNN的识别率有明显提高。     

位置正则的支持向量域描述在人脸识别中的应用研究

支持向量域描述是一种有效的一分类数据描述方法,能够有效地对单一类别的数据进行表达,并能有效地降低负样本的干扰。应用支持向量域描述方法,将人脸图像集合投影到高维特征空间构建描述特征空间中人脸图像的超球体,并定义两个超球体之间的相似性度量,应用最近邻分类器进行分类。在基于集合的人脸识别应用标准数据库上测试了该方法,在Honda/UCSD、CMU Mobo和You Tube数据分别取得100%、97.55%和59.78%的识别率。实验结果表明,该方法是一种有效的基于图像集匹配的人脸识别方法。     

Supervised parametric and non-parametric classification of chromosome images

This paper describes a fully automatic chromosome classification algorithm for Multiplex Fluorescence In Situ Hybridization (M-FISH) images using supervised parametric and non-parametric techniques. M-FISH is a recently developed chromosome imaging method in which each chromosome is labelled with 5 fluors (dyes) and a DNA stain. The classification problem is modelled as a 25-class 6-feature pixel-by-pixel classification task. The 25 classes are the 24 types of human chromosomes and the background, while the six features correspond to the brightness of the dyes at each pixel. Maximum likelihood estimation, nearest neighbor and k-nearest neighbor methods are implemented for the classification. The highest classification accuracy is achieved with the k-nearest neighbor method and k=7 is an optimal value for this classification task.     

Locally linear reconstruction for instance-based learning

Instance-based learning (IBL), so called memory-based reasoning (MBR), is a commonly used non-parametric learning algorithm. k-nearest neighbor (k-NN) learning is the most popular realization of IBL. Due to its usability and adaptability, k-NN has been successfully applied to a wide range of applications. However, in practice, one has to set important model parameters only empirically: the number of neighbors (k) and weights to those neighbors. In this paper, we propose structured ways to set these parameters, based on locally linear reconstruction (LLR). We then employed sequential minimal optimization (SMO) for solving quadratic programming step involved in LLR for classification to reduce the computational complexity. Experimental results from 11 classification and eight regression tasks were promising enough to merit further investigation: not only did LLR outperform the conventional weight allocation methods without much additional computational cost, but also LLR was found to be robust to the change of k.     

Improved k-nearest neighbor classification

k-nearest neighbor (k-NN) classification is a well-known decision rule that is widely used in pattern classification. However, the traditional implementation of this method is computationally expensive. In this paper we develop two effective techniques, namely, template condensing and preprocessing, to significantly speed up k-NN classification while maintaining the level of accuracy. Our template condensing technique aims at “sparsifying” dense homogeneous clusters of prototypes of any single class. This is implemented by iteratively eliminating patterns which exhibit high attractive capacities. Our preprocessing technique filters a large portion of prototypes which are unlikely to match against the unknown pattern. This again accelerates the classification procedure considerably, especially in cases where the dimensionality of the feature space is high. One of our case studies shows that the incorporation of these two techniques to k-NN rule achieves a seven-fold speed-up without sacrificing accuracy.     

Exact bootstrap <Emphasis Type="Italic">k</Emphasis>-nearest neighbor learners

Bootstrap aggregation, or bagging, is a method of reducing the prediction error of a statistical learner. The goal of bagging is to construct a new learner which is the expectation of the original learner with respect to the empirical distribution function. In nearly all cases, the expectation cannot be computed analytically, and bootstrap sampling is used to produce an approximation. The k-nearest neighbor learners are exceptions to this generalization, and exact bagging of many k-nearest neighbor learners is straightforward. This article presents computationally simple and fast formulae for exact bagging of k-nearest neighbor learners and extends exact bagging methods from the conventional bootstrap sampling (sampling n observations with replacement from a set of n observations) to bootstrap sub-sampling schemes (with and without replacement). In addition, a partially exact k-nearest neighbor regression learner is developed. The article also compares the prediction error associated with elementary and exact bagging k-nearest neighbor learners, and several other ensemble methods using a suite of publicly available data sets.     

结合半监督聚类和加权KNN的协同训练方法

针对协同训练方法在迭代时选择加入的无标记样本所隐含的有用信息不够,以及协同训练方法多个分类器标记不一致带来错误标记无标记样本的问题,提出了一种结合半监督聚类和加权[K]最近邻的协同训练方法。该方法在每次迭代过程中,先对训练集进行半监督聚类,选择隶属度高的无标记样本给朴素贝叶斯分类,再用加权[K]最近邻算法对多个分类器分类不一致的无标记样本重新分类。利用半监督聚类能够选择出较好表现数据空间结构的样本,而采用加权[K]最近邻算法为标记不一致的无标记样本重新标记能够解决标记不一致带来的分类精度降低问题。在UCI数据集上的对比实验验证了该算法的有效性。     

Probably correct k-nearest neighbor search in high dimensions

A novel approach for k-nearest neighbor (k-NN) searching with Euclidean metric is described. It is well known that many sophisticated algorithms cannot beat the brute-force algorithm when the dimensionality is high. In this study, a probably correct approach, in which the correct set of k-nearest neighbors is obtained in high probability, is proposed for greatly reducing the searching time. We exploit the marginal distribution of the k th nearest neighbors in low dimensions, which is estimated from the stored data (an empirical percentile approach). We analyze the basic nature of the marginal distribution and show the advantage of the implemented algorithm, which is a probabilistic variant of the partial distance searching. Its query time is sublinear in data size n, that is, O(mnδ) with δ=o(1) in n and δ≤1, for any fixed dimension m.     

基于k近邻隔离森林的异常检测

异常检测是机器学习与数据挖掘的热点研究领域之一, 主要应用于故障诊断、入侵检测、欺诈检测等领域. 当前已有很多有效的相关研究工作, 特别是基于隔离森林的异常检测方法, 但在处理高维数据时仍然存在许多困难. 提出了一种新的k近邻隔离森林的异常检算法: k-nearest neighbor based isolation forest (KNIF). 该方法采用超球体作为隔离工具, 利用第k近邻的方法来构建隔离森林, 并构建基于距离的异常值计算方法. 通过充分实验表明KNIF方法能有效地进行复杂分布环境下的异常检测, 并能适应不同分布形式的应用场景.     

面向复杂时间序列的k近邻分类器

基于时序对齐的k近邻分类器是时间序列分类的基准算法.在实际应用中,同类复杂时间序列经常展现出不同的全局特性.由于传统时序对齐方法平等对待实例特征并忽略其局部辨别特性,因此难以准确、高效地处理此类具有挑战性的时间序列.为了有效对齐并分类复杂时间序列,提出了一种具有辨别性的局部加权动态时间扭曲方法,用于发现同类复杂时间序列的共同点以及异类序列间的不同点.同时,通过迭代学习时间序列对齐点的正例集与负例集,获取每条复杂时间序列中每个特征的辨别性权重.在多个人工和真实数据集上的实验结果表明了基于局部加权对齐策略的k近邻分类器所具有的可解释性与有效性,并将所提出方法扩展至多变量时间序列分类问题中.     

A direct boosting algorithm for the k-nearest neighbor classifier via local warping of the distance metric

Though the k-nearest neighbor (k-NN) pattern classifier is an effective learning algorithm, it can result in large model sizes. To compensate, a number of variant algorithms have been developed that condense the model size of the k-NN classifier at the expense of accuracy. To increase the accuracy of these condensed models, we present a direct boosting algorithm for the k-NN classifier that creates an ensemble of models with locally modified distance weighting. An empirical study conducted on 10 standard databases from the UCI repository shows that this new Boosted k-NN algorithm has increased generalization accuracy in the majority of the datasets and never performs worse than standard k-NN.     

A hybrid medical decision making system based on principles component analysis, k-NN based weighted pre-processing and adaptive neuro-fuzzy inference system

Proper interpretation of the thyroid gland functional data is an important issue in diagnosis of thyroid disease. The primary role of the thyroid gland is to help regulation of the body's metabolism. Thyroid hormone produced by thyroid gland provides this. Production of too little thyroid hormone (hypo-thyroidism) or production of too much thyroid hormone (hyper-thyroidism) defines the types of thyroid disease. It is evident that usage of machine learning methods in disease diagnosis has been increasing gradually. In this study, diagnosis of thyroid disease, which is a very common and important disease, was conducted with such a machine learning system. In this study, we have detected on thyroid disease using principles component analysis (PCA), k-nearest neighbor (k-NN) based weighted pre-processing and adaptive neuro-fuzzy inference system (ANFIS). The proposed system has three stages. In the first stage, dimension of thyroid disease dataset that has 5 features is reduced to 2 features using principles component analysis. In the second stage, a new weighting scheme based on k-nearest neighbor (k-NN) method was utilized as a pre-processing step before the main classifier. Then, in the third stage, we have used adaptive neuro-fuzzy inference system to diagnosis of thyroid disease. We took the thyroid disease dataset used in our study from the UCI machine learning database. The obtained classification accuracy of our system was 100% and it was very promising with regard to the other classification applications in literature for this problem.