Specifying telephone systems in LOTOS

A method for detecting feature interactions in the design of telephone systems using LOTOS (language of temporal ordering specifications) is described. The specification discussed includes call forwarding on a busy line, setting a three-way call, and call waiting. The ways in which a specification of a telephone system involving such features can be constructed in LOTOS and the ways in which LOTOS tools can be used to detect unwanted interactions of the features are illustrated using simple examples. The techniques used are step-by-step execution, symbolic execution, and compositional execution with test processes     

基于着色Petri网的电信业务特性集成模型

“业务特性”的快速集成是电信系统发展的趋势,本文提出了基于着色Ｐｅｔｒｉ网（ＣＰ网）的业务特性集成模型。在本方法中,基本的电话系统以及每个业务特性以一个ＣＰ网来模拟,当把一个业务特性集成到系统中时,相应于把代表它的ＣＰ网嵌入系统的网中。本模型既形象化地模拟业务特性的集成,同时可以用数学方法来分析业务特性的行为,可应用于检测和解决业务特性的交互作用。     

基于邻域像素差分矩阵的彩色空域图像隐写分析特征

通过分析灰度空域图像隐写直接扩展至彩色空域图像潜在的安全问题,提出了针对彩色空域图像隐写的隐写分析特征。首先,计算颜色通道内邻域像素差分矩阵的Markov转移概率矩阵,提取颜色通道内特征并进行融合;然后,计算颜色通道间邻域像素差分矩阵的Markov转移概率矩阵,并将水平和垂直共4个方向与对角和反对角共4个方向的转移概率矩阵分别进行融合,作为颜色通道间特征;最后,颜色通道内特征与颜色通道间特征构成最终的隐写分析特征,并使用集成分类器作为特征训练和测试工具。实验结果表明,该隐写分析特征可有效检测彩色空域图像隐写,并且颜色通道间特征能有效捕获内容自适应隐写对彩色图像的颜色通道间相关性的影响。     

基于相关向量机的网络入侵检测算法

针对支持向量机理论中存在的问题：训练样本数量多以及必须满足MerCer条件等,提出了一种基于相关向量机（RVM）的网络入侵检测方法。首先采用“删除特征”法对KDD99数据集中的41个特征进行评级,筛选出针对不同入侵类型的重要特征和非重要特征,然后只选择重要特征进行匹配。结果表明,这种方法与基于支持向量机（SVM）的入侵检测模型相比,具有更高的检测率和更低的误警率。     

A sparse representation method for image-based surface defect detection

In this paper, an efficient sparse representation-based method is presented for detecting surface defects. The proposed method uses the sparse degree of coefficient in the redundant dictionary for checking whether the test image is defective or not, and the binary representation of the defective images is obtained, according to the global coefficient feature. Owing to the requirements for the efficiency and detecting quality, the block proximal gradient operator is introduced to speed up the online dictionary learning. Considering the correlation among the testing samples, prior knowledge is applied in the orthogonal-matching-pursuit sparse representation algorithm to improve the speed of sparse coding. Experimental results demonstrate that the proposed detection method can effectively detect and extract the defects of the surface images, and has broad applicability.     

基于多特征图像显著目标的自动识别

现有的许多显著目标检测算法,大都依赖像素间的相互关系,而缺乏对焦点对象特征的理解.本文提出了一种无监督的显著目标自动识别算法.首先,应用具有仿生学特质的多尺度Gabor模型检测图像中的初级特征,构成显著目标的基本元素特征.在此基础上,结合显著目标的自身特性定义图像的局部特征,进一步确定感兴趣区域的位置.最后,在模糊阈值算法的基础上提出一种新的提取策略,并将其应用到由初级特征和局部特征构成的显著图上,从而准确地确定显著目标的位置.将此方法应用于具有不同特点的图像进行仿真实验,得到了较好的结果,证明该算法是切实可行的.     

广告段落分割系统中的字幕检测

分析了视频节目中字幕视频的特点,提出了一种基于Sobel算子边缘检测的字幕检测方法,对实验结果进行分析,实验结果表明,该方法可有效检测视频序列中的字幕片段。     

A perceptual scheme for fully automatic video shot boundary detection

In this paper, we propose a novel and robust modus operandi for fast and accurate shot boundary detection where the whole design philosophy is based on human perceptual rules and the well-known “Information Seeking Mantra”. By adopting a top–down approach, redundant video processing is avoided and furthermore elegant shot boundary detection accuracy is obtained under significantly low computational costs. Objects within shots are detected via local image features and used for revealing visual discontinuities among shots. The proposed method can be used for detecting all types of gradual transitions as well as abrupt changes. Another important feature is that the proposed method is fully generic, which can be applied to any video content without requiring any training or tuning in advance. Furthermore, it allows a user interaction to direct the SBD process to the user's “Region of Interest” or to stop it once satisfactory results are obtained. Experimental results demonstrate that the proposed algorithm achieves superior computational times compared to the state-of-art methods without sacrificing performance.     

基于PCANet和SVM的谎言测试研究

主成分分析网络（Principal Component Analysis Network,PCANet）是基于深度学习理论的一种非监督式的特征提取方法,它克服了手工提取特征的缺点,目前其有效性仅仅在图像处理领域中得到了验证。本文针对当前谎言测试方法中脑电信号特征提取困难的缺点,首次将PCANet方法应用到一维信号的特征提取领域,并对测谎实验的原始脑电信号提取特征,然后使用支持向量机（Support Vector Machine,SVM）将说谎者和诚实者的两类信号进行分类识别,将实验结果和其它分类器及未使用特征提取的分类效果进行了比较。实验结果显示相对未抽取任何特征的方法,提出的方法PCANet－SVM可以获得更高的训练和测试准确率,表明了PCANet方法对于脑电信号特征提取的有效性,也为基于脑电信号的测谎提供了一种新的途径。     

视频检索中基于GMM聚类的无监督情感场景检测

为了高效地从视频中检索出激动人心的场面,提出了一种基于高斯混合模型的无监督情感场景检测方法.首先,从面部选取42个特征点,并定义10种面部特征;然后,利用高斯混合模型将视频的帧划分为多个聚类;最后,利用每一帧的面部表情分类结果将情感场景划分为单个聚类,并通过场景集成和删除完成检测.在生活记录视频和MMI人脸表情数据库上的实验结果表明,该方法的检测率、分类率分别高达98％,95％,检测5分钟左右的情感场景视频仅需0.138 s,性能优于几种较为先进的检测方法.     

基于兴趣区检测的地面目标识别方法研究

提出了一种基于兴趣区检测的地面目标识别方法。根据目标模型的有关几何形状和热辐射知识,采用自上而下的计算模型,计算所提取的基元特征的显著性,并根据基元特征的显著性的大小确定出注视点及相应的兴趣区,然后对兴趣区中的图像进行分割,并利用模型知识对分割图像进行识别。该方法对已有的红外目标图像系列取得了很好的识别效果,提高了识别精度和弱小目标检测识别的能力。     

基于DBN 的频率与相位编码信号快速调制识别方法

为提升低信噪比条件下雷达/ 通信频率、相位编码信号调制识别性能,降低特征提取复杂度,提出了基于深度信念网络DBN(Deep Belief Network, DBN)以及快速特征提取的调制识别方法。结合快速傅里叶累加算法FAM(FFT Accumulation Method)算法,提出了将循环谱估计图像转化为有效可识别特征向量的提取算法;设计了用于编码信号调制识别的DBN 网络训练与识别框架。仿真结果表明,文中方法较传统方法具有更低的特征提取与预处理复杂度,提取的特征在几种典型编码调制模式信号中具有明显区分,DBN 训练识别框架对雷达/ 通信编码信号调制识别均具有可行性与有效性,在低信噪比条件下对无线电编码信号有更高的识别正确率。     

Using Feature-Based Models with Complexity Penalization for Selecting Features

Feature selection and inference through modeling are combined into one method based on a network that can be used to point out irrelevant, redundant and dependent features in the data. It is shown that this network method is efficient in terms of reducing the number of calculations for estimating the probabilities under different model assumptions by breaking the data into fractions. We prove that the probability estimations within the network method lead to the detection of non-informative features with probability one if the data is sufficiently large. The proposed method’s accuracy in detecting complex relations between features, selecting informative features and classifying data-sets with different dimensions is assessed through experiments using both synthetic and real data. The results from the network method compare favorably with those from the well-known and powerful feature selection algorithms. It is further shown that the network method can handle complex relations between the features that are intractable for other algorithms.     

基于Windows主机的未知蠕虫主动检测系统

相对于传统的反病毒软件,本研究提出了一种新的基于主机的蠕虫检测系统.这种新的检测方法通过分析监测已知蠕虫病毒对计算机性能参数造成的影响,对未知的蠕虫进行判决检测,达到网络预警的效果.我们通过监测主机的323个系统特征计数器以反应计算机的性能特征,并利用自行设计的对比特征选择系统对原始数据进行预处理.本研究采用贝叶斯网络分类算法对带有标签的各训练数据子集进行分类训练,产生用以判决未知蠕虫的分类判决规则.在模拟计算机各种应用状态下,我们在搭建的实验局域网上对这种新的监测系统进行测试评估.本研究提出的检测系统对采集的未知网络蠕虫达到80%以上的判决准确率,对已知蠕虫有着99%以上的检测准确率,有着很高的实用性和推广性.     

结合多特征融合和极限学习机的红外图像目标分类方法

针对红外图像目标分类问题,提出了结合多特征融合和极限学习机（extreme learning machine,ELM）的方法。采用主成分分析（principal component analysis,PCA）、局部二值模式（local binary pattern,LBP）以及尺度不变特征变换（scale-invariant feature transform,SIFT）三类特征分别描述红外图像中目标的像素分布、局部纹理以及特征点信息。三类特征从不同侧面反映红外图像目标特性,因此具有互为补充的优势。在此基础上,基于多重集典型相关分析（multiset canonical correlations analysis,MCCA）对三类特征进行融合处理,获得统一的特征矢量。融合后的特征不仅继承了原始三类特征的鉴别特性,还有效去除了冗余信息。分类过程中,采用极限学习机作为基础分类器对融合特征矢量进行分类。极限学习机具有参数少、效率高、精度高和稳健性强等显著特点,有利于提高红外目标分类的整体性能。因此,所提出的方法通过结合多特征和极限学习机的优势综合提升了目标识别性能。在实验过程中,采用四类飞机目标的红外图像对所提出方法进行了性能测试。根据与现有几类方法的对比,实验结果证明了提出方法的性能优势。     

平面旋转人脸检测与特征定位方法研究

提出了一种基于角点检测、AdaBoost算法和C-V方法的平面旋转人脸检测及特征定位方法.方法首先根据AdaBoost算法训练样本得到脸、眼、鼻、嘴4个检测器;然后以角点作为眼睛的候选点,枚举任意两个角点构造可能的人脸区域,并在区域内运用人脸检测器进行检测;接着利用眼、鼻、嘴检测器检测出人脸特征所在的矩形区域;最后利用C-V方法从各个特征区域中分割出人脸特征的轮廓,进而得到人脸关键特征点的位置.在CMU平面旋转测试集上的检测率为94.6%,误报24个,提取出的特征点位置准确.实验结果表明方法是有效的.     

基于稀疏自动编码网络的水声通信信号调制识别

研究了基于稀疏自动编码网络的水声通信信号识别方法。首先利用稀疏自动编码网络对接收信号的功率谱识别分类,得到除PSK外信号的调制类型,然后对识别结果为PSK的信号做四次方谱,最后利用稀疏自动编码网络完成对QPSK和8PSK的识别分类。仿真实验表明,稀疏自动编码网络能从接收信号的谱信息中自动提取有效谱特征。与传统基于功率谱特征提取的识别方法相比,本文算法减少了依赖领域知识的特征提取环节,识别性能优于传统算法。      

基于分形曲面尺度斜率特征的弱小目标检测

鉴于弱小目标检测所固有的难点及常用的检测方法不能准确、稳定地检测出目标,提出了运用分形曲面尺度斜率特征检测弱小目标的方法。通过实际数据分析可以得出:相比常用的分形维数和分形拟合误差等检测特征,分形曲面尺度斜率特征在表征人造目标与自然背景的差异上更加明显,在抗图像噪声干扰上也更为优异,有着更强的鲁棒性。该方法普遍适用于检测自然环境中的弱小目标,尤其在对空弱小目标方面,检测概率更高。无论背景、飞行姿态、目标类型发生怎样的变化,经本文算法运算后只需一步简易的分割就可以检测出微弱暗小目标。     

Learning discriminative representations for multi-label image recognition

Multi-label recognition is a fundamental, and yet is a challenging task in computer vision. Recently, deep learning models have achieved great progress towards learning discriminative features from input images. However, conventional approaches are unable to model the inter-class discrepancies among features in multi-label images, since they are designed to work for image-level feature discrimination. In this paper, we propose a unified deep network to learn discriminative features for the multi-label task. Given a multi-label image, the proposed method first disentangles features corresponding to different classes. Then, it discriminates between these classes via increasing the inter-class distance while decreasing the intra-class differences in the output space. By regularizing the whole network with the proposed loss, the performance of applying the well-known ResNet-101 is improved significantly. Extensive experiments have been performed on COCO-2014, VOC2007 and VOC2012 datasets, which demonstrate that the proposed method outperforms state-of-the-art approaches by a significant margin of 3.5% on large-scale COCO dataset. Moreover, analysis of the discriminative feature learning approach shows that it can be plugged into various types of multi-label methods as a general module.