新颖人脸检测的核方法

核主成分分析（kernel PCA）是PCA的非线性扩展。该研究将kernel PCA应用于新颖人脸检测。作为训练数据的人脸图像被映射到高维特征空间。在特征空间中,kernel PCA抽取数据分布的主成分,构成主子空间。在该子空间中,通过优化方法寻找包含训练数据的最小超平面,作为新颖检测的决策界面。该新方法在ORL人脸图像库的数据集中进行了实验,测试精度较线性PCA方法有所提高。     

Evaluation of novelty metrics for sentence-level novelty mining

This work addresses the problem of detecting novel sentences from an incoming stream of text data, by studying the performance of different novelty metrics, and proposing a mixed metric that is able to adapt to different performance requirements. Existing novelty metrics can be divided into two types, symmetric and asymmetric, based on whether the ordering of sentences is taken into account. After a comparative study of several different novelty metrics, we observe complementary behavior in the two types of metrics. This finding motivates a new framework of novelty measurement, i.e. the mixture of both symmetric and asymmetric metrics. This new framework of novelty measurement performs superiorly under different performance requirements varying from high-precision to high-recall as well as for data with different percentages of novel sentences. Because it does not require any prior information, the new metric is very suitable for real-time knowledge base applications such as novelty mining systems where no training data is available beforehand.     

Information theoretic novelty detection

We present a novel approach to online change detection problems when the training sample size is small. The proposed approach is based on estimating the expected information content of a new data point and allows an accurate control of the false positive rate even for small data sets. In the case of the Gaussian distribution, our approach is analytically tractable and closely related to classical statistical tests. We then propose an approximation scheme to extend our approach to the case of the mixture of Gaussians. We evaluate extensively our approach on synthetic data and on three real benchmark data sets. The experimental validation shows that our method maintains a good overall accuracy, but significantly improves the control over the false positive rate.     

On-line expectation-based novelty detection for mobile robots

This paper presents a recurrent neural network based novelty filter where a Scitos G5 mobile robot explored the environment and built dynamic models of observed sensory–motor values, then the acquired models of normality are used to predict the expected future values of sensory–motor inputs during patrol. Novelties could be detected whenever the prediction error between models-predicted values and actual observed values exceeded a local novelty threshold. The network is trained on-line; it grows by inserting new nodes when abnormal observation is perceived from the environment; and also shrinks when the learned information is not necessary anymore. In addition, the network is also capable of learning region-specific novelty thresholds on-line continuously.To evaluate the proposed algorithm, real-world robotic experiments were conducted by fusing sensory perceptions (vision and laser sensors) and the robot motor control outputs (translational and rotational velocities). Experimental results showed that all of the novelty cases were highlighted by the proposed algorithms and it produced reliable local novelty thresholds while the robot patrols in the noisy environment. The statistical analysis showed that there was a strong correlation between the novelty filter responses and the actual novelty status. Furthermore, the filter was also compared with another novelty filter and the results showed that the proposed system performed better novelty detection.     

Machine learning-based novelty detection for faulty wafer detection in semiconductor manufacturing

Since semiconductor manufacturing consists of hundreds of processes, a faulty wafer detection system, which allows for earlier detection of faulty wafers, is required. statistical process control (SPC) and virtual metrology (VM) have been used to detect faulty wafers. However, there are some limitations in that SPC requires linear, unimodal and single variable data and VM underestimates the deviations of predictors. In this paper, seven different machine learning-based novelty detection methods were employed to detect faulty wafers. The models were trained with Fault Detection and Classification (FDC) data to detect wafers having faulty metrology values. The real world semiconductor manufacturing data collected from a semiconductor fab were tested. Since the real world data have more than 150 input variables, we employed three different dimensionality reduction methods. The experimental results showed a high True Positive Rate (TPR). These results are promising enough to warrant further study.     

Kernel ellipsoidal trimming

Ellipsoid estimation is important in many practical areas such as control, system identification, visual/audio tracking, experimental design, data mining, robust statistics and statistical outlier or novelty detection. A new method, called kernel minimum volume covering ellipsoid (KMVCE) estimation, that finds an ellipsoid in a kernel-defined feature space is presented. Although the method is very general and can be applied to many of the aforementioned problems, the main focus is on the problem of statistical novelty/outlier detection. A simple iterative algorithm based on Mahalanobis-type distances in the kernel-defined feature space is proposed for practical implementation. The probability that a non-outlier is misidentified by our algorithms is analyzed using bounds based on Rademacher complexity. The KMVCE method performs very well on a set of real-life and simulated datasets, when compared with standard kernel-based novelty detection methods.     

Online face recognition system based on multi-view algorithm and Kernel PCA

An online face recognition system is presented in the paper. To online face recognition system, we should consider the recognition rate, the image compression and image size. In the paper we researched the innovation technologies for face recognition system, including Kernel Principal Component Analysis （Kernel PCA）, Delta low-pass wavelet filter, and face recognition algorithm using multiple images. Kernel PCA is derived to classify the characteristics of training images in the database. Delta low-pass wavelet filter is used to reduce the image size. A face recognition algorithm using multiple images is presented to improve the recognition rate. Simulation experiment shows that in the case of packet loss recognition rate is improved highly.     

Feature extraction for novelty detection as applied to fault detection in machinery

Novelty detection is a pattern recognition technique used when there is one well characterized normal state and the abnormal (or novel) states are poorly described because of lack of data. Data deficiency of these states may arise due to cost and difficulty in measuring them - e.g. failed equipment states in equipment health monitoring. Normal pattern recognition techniques have a wide array of methods for reducing the number of features initially employed to characterize classes. These techniques are of limited use in novelty detection primarily because they are focused on representing the data accurately in a subspace rather than on finding a subspace where the classes can easily be discriminated, or they are optimized to distinguish between all classes rather than on focusing on distinguishing solely between normal and abnormal classes. The proposed methodology will enable feature extraction in unbalanced classification problems where the well-sampled normal data is expected to be orbited by the under-sampled fault data. The technique will be demonstrated to work well with artificial and actual machinery health data.     

基于PCA初始化卷积核的CNN手写数字识别算法

针对卷积神经网络对手写数字识别训练在卷积核随机初始化情况下收敛速度慢和识别率低的问题，提出一种主成分分析（PCA）初始化卷积核的卷积神经网络（CNN）手写数字识别算法。算法首先选取训练样本集并将其送入CNN，在相应层对Feature Map进行全覆盖取图像块处理，然后进行分层PCA学习，将学习到的特征向量做为对应卷积层的卷积核参数进行初始化，最后再用这些卷积核对原始图像进行卷积操作。实验结果表明，与随机初始化卷积核的CNN手写数字识别算法相比，改进的算法在应用MNIST数据库训练时不仅收敛，而且在产生相同均方误差的情况下迭代次数少，识别率高。     

Intrinsically motivated learning systems based on biologically-inspired novelty detection

Intrinsic motivations play an important role in human learning, particularly in the early stages of childhood development, and ideas from this research field have influenced robotic learning and adaptability. In this paper we investigate one specific type of intrinsic motivation, that of novelty detection and we discuss the reasons that make it a powerful facility for continuous learning. We formulate and present one original type of biologically inspired novelty detection architecture and implement it on a robotic system engaged in a perceptual classification task. The results of real-world robot experiments we conducted show how this original architecture conforms to behavioural observations and demonstrate its effectiveness in terms of focusing the system’s attention in areas that are potential for effective learning.     

基于全局网络PCA的DDoS攻击检测方法

随着网络规模的不断扩充,对于DDoS攻击的集中式检测方法已经无法满足实时性和准确性等要求。针对大规模网络中的DDoS攻击行为,提出了一种基于全局PCA的分布式拒绝服务攻击检测方法(WPCAD)。该方法由传统的OD矩阵得出各节点的ODin矩阵,各分布式处理单元通过PCA分析到达该节点的多路OD流之间的相关性,利用DDoS攻击流引起流量之间相关性突变的特性来完成检测。该方法采用分布式处理的方式,降低了检测数据所消耗的带宽,并满足了检测的实时性。实验结果表明该方法具有更好的检测效果。     

An improved weighted recursive PCA algorithm for adaptive fault detection

A novel weighted adaptive recursive fault detection technique based on Principal Component Analysis (PCA) is proposed to address the issue of the increment in false alarm rate in process monitoring schemes due to the natural, slow and normal process changes (aging), which often occurs in real processes. It has been named as weighted adaptive recursive PCA (WARP).The aforementioned problem is addressed recursively by updating the eigenstructure (eigenvalues and eigenvectors) of the statistical detection model when the false alarm rate increases given the awareness of non-faulty condition. The update is carried out by incorporating the new available information within a specific online process dataset, instead of keeping a fixed statistical model such as conventional PCA does. To achieve this recursive updating, equations for means, standard deviations, covariance matrix, eigenvalues and eigenvectors are developed. The statistical thresholds and the number of principal components are updated as well.A comparison between the proposed algorithm and other recursive PCA-based algorithms is carried out in terms of false alarm rate, misdetection rate, detection delay and its computational complexity. WARP features a significant reduction of the computational complexity while maintaining a similar performance on false alarm rate, misdetection rate and detection delay compared to that of the other existing PCA-based recursive algorithms. The computational complexity is assessed in terms of the Floating Operation Points (FLOPs) needed to carry out the update.     

一种基于核距离的核函数度量方法

核方法的效果依赖于所使用的核,因此核的选择和其参数的确定是至关重要的。从特定的数据中学习核需要核度量方法评价核的质量。核排列度量核与学习任务的一致性,因为它具有高效性和有效性,是目前应用最为广泛的核度量方法。然而,有研究表明,核排列仅是最优核函数的充分非必要条件。其主要原因是核排列在特征空间中不具有线性变换不变性。提出了一种新的核度量方法用于核选择,称其为核距离排列。该方法能够克服核排列的局限性,并且同样具有高效性和简单的形式。对比实验表明,该方法能够有效地对核进行度量。     

基于核中心距比值法的入侵检测

为了解决目前许多推广能力强的入侵检测系统训练时间长的问题,提出了一种不需要花费时间训练,同时又具有推广能力的入侵检测方法。该文利用核方法,巧妙地对非线性可分的入侵检测数据进行分类,提出了利用核中心距离比值法进行入侵检测的框图。用系统调用执行迹数据仿真的结果表明,核中心距比值分类法不但节省训练时间,而且检测性能比较理想。     

基于多通道PCA模型的手写汉字识别方法

为了提高手写汉字的识别率和降低训练时间，提出了一种基于多通道PCA(Principal component analysis)模型的手写汉字识别方法．该方法首先根据汉字的结构特点，将手写汉字分解为“一”、“I”、“J”、“＼”4种方向子模式，然后分别对每个子模式进行主分量分析，最后通过建立起每类汉字的多通道PCA模型来进行手写汉字的识别．该方法既兼顾了主分量对手写汉字的描述能力，又有效地降低了建立模型的训练时间．针对1034类别的手写汉字样本的实验结果表明，该汉字识别方法的识别率较欧氏距离分类器提高了4．4个百分点，而其训练时间则明显低于直接进行PCA重建的识别方法，由此可见，该方法是有效的。     

融合Kernel PCA形状先验信息的变分图像分割模型

目的 基于能量最小化的变分图像分割方法已经受到研究人员的广泛重视,取得了丰硕成果。但是,针对图像中存在的噪音污染、目标被遮挡等情况,则难以正确分割。引入先验形状信息是解决该问题的一个重要方向,但是随之而带来的姿态变化问题是一个难点。传统的做法是在每步迭代过程中单独计算姿态变换参数,导致计算量大。方法 在基于Kernel PCA(KPCA)的形状先验模型基础上,提出一种具有内在的姿态不变性的KPCA形状先验模型,并将之融合到C-V变分图像分割模型中。结果 提出模型无须在每步迭代中显式地单独计算姿态变换参数,相对于C-V模型分割正确率能够提高7.47%。同时,针对KPCA模型中计算高斯核函数的参数σ取值问题,也给出一种自适应的计算方法。结论 理论分析及实验表明该模型能较好地解决先验形状与目标间存在的仿射变化问题,以及噪音、目标被遮挡等问题。     

Palmprint Recognition by Applying Wavelet-Based Kernel PCA

This paper presents a wavelet-based kernel Principal Component Analysis (PCA) method by integrating the Daubechies wavelet representation of palm images and the kernel PCA method for palmprint recognition. Kernel PCA is a technique for nonlinear dimension reduction of data with an underlying nonlinear spatial structure. The intensity values of the palmprint image are first normalized by using mean and standard deviation. The palmprint is then transformed into the wavelet domain to decompose palm images and the lowest resolution subband coeffcients are chosen for palm representation. The kernel PCA method is then applied to extract non-linear features from the subband coeffcients. Finally, similarity measurement is accomplished by using weighted Euclidean linear distance-based nearest neighbor classifier. Experimental results on PolyU Palmprint Databases demonstrate that the proposed approach achieves highly competitive performance with respect to the published palmprint recognition approaches.     

A black hole novelty detector for video analysis

A black hole is a region of space that has so much mass concentrated in it that there is no way training data for a nearby object to escape its gravitational pull. In this paper we are inspired by this phenomenon to create a new form of novelty detector. We consider the of a given class as a black hole. For multi-class data we are dealing with multiple black holes. A test point is pulled by the centroids of different black holes as well as its K nearest neighbours. The gravitational pull is modelled as an iterative process, where the forces acting on a point are constantly changing with time as the test point moves in multi-dimensional space corresponding to these forces. Once the algorithm has converged, a thresholding scheme is applied to determine whether the test point has been pulled within the boundary of the black hole or not. Any points that lie outside all known black holes are deemed to be novel. We compare this novelty detector with other well-known models of novelty detection on a video analysis application and show very promising results.     

概率核主成分分析及其应用

主成分分析（PCA）、核主成分分析（KPCA）和概率主成分分析（PPCA）是已经取得广泛应用的特征提取方法。提出一种基于概率核主成分分析（PKPCA）的检测液晶屏幕亮点的方法。作为对PPCA的一种非线性扩展,PKPCA在PPCA的基础上引入了核函数方法,因而其捕获模式非线性特征的能力更强。在KPCA和PPCA的基础上推导了PKPCA过程公式,并在检测液晶屏幕亮点的应用中将PKPCA、PPCA、PCA算法进行比较。实验结果表明,PKPCA的检测率和局部信噪比优于其他两者。