相关向量机多分类算法的研究与应用
|
摘要
相关向量机(Relevance Vector Machine, RVM)是贝叶斯统计学习理论(StatisticalLearing Theory,SLT)发展的产物，是一种有监督机器学习的模式识别新方法。该方法由支持向量机(Support Vector Machine, SVM)理论演变而来，相比后者，具有解更稀疏、核函数选择更自由、泛化能力更强、鲁棒性更好等优点，在小样本的统计学习问题中的表现尤其突出，近几年已经在应用领域得到了快速发展，在模式分类、故障诊断、智能预测、语音及图像信息处理等方面均有很好的表现。但是，相关向量机在解决多类模式识别问题时，由于计算过程比较复杂，仍存在分类精度与训练识别时间无法兼顾的矛盾。
本课题针对RVM算法存在的不足，对算法的结构和关键步骤，如核函数的选择、分类器的设计以及控制参数的调整等进行了深入研究和大量实验仿真工作，并对多分类问题中应用最广泛且分类精度最高的“一对一”分类器进行了改进。改进后的多分类方法，在基本保持原有的分类精度的基础上，大幅度提升了算法在类别数较多的模式识别问题上的分类时间，使RVM算法应用的实时性有了明显的提高。此外，从应用角度出发，将改进后的RVM算法应用于人脸识别及汽车发动机失火故障检测等问题的模式识别中，均取得了良好的效果。
首先，详细论述相关向量机的研究现况和基本理论,并且提出相关向量机中仍需解决的关键问题。为了提高相关向量机学习算法在多模式识别中的分类速度，对相关向量机多分类方法进行了分析和研究，发现比较次数过多是该方法计算量大的主要原因。提出了一种在每轮比较中，排除最差类别的新方法。该方法使比较次数逐级减少，并且当类别数较多时，总计算量减少尤其明显。通过仿真实验说明了该方法的有效性,对数据分类的实验结果表明，新方法与传统分类器相比，在基本不影响分类正确率的前提下，机器训练与识别次数显著减少，算法运行速度明显提高。
其次，为了解决人脸识别问题中对准确性、实时性、稳定性的要求，对传统的人脸识别方法进行了研究，提出一种基于改进相关向量机的人脸识别方法。文章利用小波变换对人脸图像进行预处理；根据PCA方法对处理后的人脸图像进行特征提取；利用相关向量机多分类模型进行人脸识别。与基于SVM的人脸识别方法进行比较，结果表明RVM具有高于SVM的鲁棒性，人脸识别的正确率更高、实时性好、可靠性更强。
再次，当人脸图像含有较多噪声时，识别正确率会有很明显的下降。目前的人脸识别技术对此问题尚无较好的解决办法。本文提出一种采用相关向量机的人脸识别方法，利用机器学习对小波分解和PCA变换后的人脸数据库样本进行训练，得到的相关向量构成“超平面”作为差异样本的分类面，并利用改进的“一对一”方法实现多类别模式识别。对加噪声的识别对象进行了大量的仿真实验结果表明，与传统方法相比，新方法对图像噪声不敏感，具有更高的识别率和很强的鲁棒性。另外，对拍照光线、角度变化、物体遮挡、分辨率不足等条件下的人脸图像识别，也采用新方法进行了实验分析和讨论。
最后，应用RVM算法研究汽车发动机故障诊断问题。研究发现算法中的惩罚因子和径向基核函数参数对分类准确率有着很大的影响，本文结合粒子群（PSO）算法对参数进行优化，并把该优化算法用于汽车发动机故障诊断中。针对样本的特征参数会随发动机转速变化的问题，提出了一种超参数自适应拟合的增量学习方法。在发动机失火故障诊断中，建立汽车尾气中各气体的体积分数与失火故障原因的映射关系，并对不同档位归一化处理的数据进行增量机器训练，对得到的超参数进行非线性拟合，并将训练好的RVM模型应用于故障分类诊断。仿真实验表明新方法不仅诊断结果准确可靠，而且解决了传统方法实现变速动态检测的困难。
在论文的结尾，对课题的研究工作进行了总结，并对进一步研究工作进行了展望。
Relevance vector machine (RVM) technique is a novel pattern recognition method ofsupervision machine learning which is based on Bayesian learning theory. It was developedon the basis of Support vector machine(SVM) learning theory, compared with the SVM, it hasthe benefits of sparser model、the facility to utilize arbitrary kernel functions、more accurate、strong robusticity、 intensity generalization ability， and so on. RVM algorithm has beendeveloped rapidly in the fields of application, and proved better in pattern classification、faultdiagnosis、intelligent forecast、information processing of voice and image, etc. Whereas，tosolve the questions of multi-mode pattern recognition，RVM algorithm still suffer from theproblems of looking after both sides of accuracy and real-time because of the complexity ofcomputational process.
According to the insufficiency of RVM, the structure and key steps of the algorithm,including kernel functions selection, classifier construct, and control parameter setting, aredeeply investigated in this paper and improved on “one against one” classifier which has beenthe highest accuracy and broad application are proposed to improve. The improvedclassification method proposed is utilized successfully to reduce the time of classification andincrease in real-time without cutting down the accuracy. In addition, improved algorithm inthis paper is applied in face recognition&automobile engine fault diagnosis, and behavedwell.
Firstly, the research situation and the fundamental theory about RVM is detail discussedin this paper, and the solved key problem of RVM is presented. In order to improveclassification speed of multiclass pattern recognition based on relevance vector machinelearning algorithm， investigated the method of relevance vector machine algorithm inmulti-mode classification, and found that the comparison too many times was the main reasonfor large amount of calculation. Proposed a new waythat eliminated the most dissimilar classin each round of comparison. Comparison times were reduced step by step per cycle, and theclassification was more, the decrease in the total calculation amount was more obvious. Thevalidity of this method has been proved by some simulated examples, and the experimentalresults of data classification show that compared with traditional classifier, the training timesand the recognition times of the novel method are greatly reduced under the premise of hardly influence classification accuracy, the algorithm running speed is improved obviously.
Secondly, to solve the problems existed in face recognition, such as lack of accuracy、real-time and stability，a new face recognition approach based on improved relevance vectormachine is presented in this paper. Firstly, the wavelet transform is applied to preprocess faceimage to reduce the impact of expression change. Then, in order to extract key features of theprocessed face image, use the principal component analysis (PCA) method. Finally, the RVMclassification model is adopted for identifying. In comparison with the support vectormachine(SVM) method, the RVM approach performs well and can obtain more satisfactoryresults in terms of recognition rates、real-time and reliability.
Thirdly, the face images recognition accuracy will be obvious decline when the objectscontain more noise. At present, face recognition technology to solve this problem is no betterway. In this paper, a new method of face recognition based on relevance vector machine waspresented. After the wavelet decomposition and PCA transform, relevance vectors fromsample training constituted a "hyperplane" as the differences in the classification of thesamples by machine learning algorithm and used the improved “one against one” method toachieve multi-class pattern recognition. Compared with the former method, a large number ofsimulation results show that the new method used in noisy objects being recognized is notsensitive to image noise, with a more accurate and strong robusticity. In addition, photo light、angle change、occlusion、low resolution ratio and so on, also be discussed and analysed inexperiment with new method in this paper.
Finally, application of RVM in the automobile engine fault diagnosis is investigated.From the study we know that the parameter of penalty factor and kernel paraeter play a veryimportant roal on the diagnosis model, so the Particle Swarm Optimization(PSO) is used tooptimize the parameters, this algorithms practical applied to automotive engine fault diagnosis.Considering the problem of the variation of characteristic parameters are followed by enginerotational speed, puts forward a adaptive fitting of super-parameters on incremental learning.To the problem of engine misfire, mapping relations established between gas volume fractionand the cause of the misfire, used normalized data with different gears in machine training,ajusted super-parameters by curve fitting, and the trained RVM model applied in faultclassification and diagnosis.The simulation experiments shows the results of new method isnot only accurate and reliable but also resolve the problem of dynamic detection with variable speed in traditional methods.
In conclusion, a simple summary is made and some research aspects are presented in thefuture.
本课题针对RVM算法存在的不足，对算法的结构和关键步骤，如核函数的选择、分类器的设计以及控制参数的调整等进行了深入研究和大量实验仿真工作，并对多分类问题中应用最广泛且分类精度最高的“一对一”分类器进行了改进。改进后的多分类方法，在基本保持原有的分类精度的基础上，大幅度提升了算法在类别数较多的模式识别问题上的分类时间，使RVM算法应用的实时性有了明显的提高。此外，从应用角度出发，将改进后的RVM算法应用于人脸识别及汽车发动机失火故障检测等问题的模式识别中，均取得了良好的效果。
首先，详细论述相关向量机的研究现况和基本理论,并且提出相关向量机中仍需解决的关键问题。为了提高相关向量机学习算法在多模式识别中的分类速度，对相关向量机多分类方法进行了分析和研究，发现比较次数过多是该方法计算量大的主要原因。提出了一种在每轮比较中，排除最差类别的新方法。该方法使比较次数逐级减少，并且当类别数较多时，总计算量减少尤其明显。通过仿真实验说明了该方法的有效性,对数据分类的实验结果表明，新方法与传统分类器相比，在基本不影响分类正确率的前提下，机器训练与识别次数显著减少，算法运行速度明显提高。
其次，为了解决人脸识别问题中对准确性、实时性、稳定性的要求，对传统的人脸识别方法进行了研究，提出一种基于改进相关向量机的人脸识别方法。文章利用小波变换对人脸图像进行预处理；根据PCA方法对处理后的人脸图像进行特征提取；利用相关向量机多分类模型进行人脸识别。与基于SVM的人脸识别方法进行比较，结果表明RVM具有高于SVM的鲁棒性，人脸识别的正确率更高、实时性好、可靠性更强。
再次，当人脸图像含有较多噪声时，识别正确率会有很明显的下降。目前的人脸识别技术对此问题尚无较好的解决办法。本文提出一种采用相关向量机的人脸识别方法，利用机器学习对小波分解和PCA变换后的人脸数据库样本进行训练，得到的相关向量构成“超平面”作为差异样本的分类面，并利用改进的“一对一”方法实现多类别模式识别。对加噪声的识别对象进行了大量的仿真实验结果表明，与传统方法相比，新方法对图像噪声不敏感，具有更高的识别率和很强的鲁棒性。另外，对拍照光线、角度变化、物体遮挡、分辨率不足等条件下的人脸图像识别，也采用新方法进行了实验分析和讨论。
最后，应用RVM算法研究汽车发动机故障诊断问题。研究发现算法中的惩罚因子和径向基核函数参数对分类准确率有着很大的影响，本文结合粒子群（PSO）算法对参数进行优化，并把该优化算法用于汽车发动机故障诊断中。针对样本的特征参数会随发动机转速变化的问题，提出了一种超参数自适应拟合的增量学习方法。在发动机失火故障诊断中，建立汽车尾气中各气体的体积分数与失火故障原因的映射关系，并对不同档位归一化处理的数据进行增量机器训练，对得到的超参数进行非线性拟合，并将训练好的RVM模型应用于故障分类诊断。仿真实验表明新方法不仅诊断结果准确可靠，而且解决了传统方法实现变速动态检测的困难。
在论文的结尾，对课题的研究工作进行了总结，并对进一步研究工作进行了展望。
Relevance vector machine (RVM) technique is a novel pattern recognition method ofsupervision machine learning which is based on Bayesian learning theory. It was developedon the basis of Support vector machine(SVM) learning theory, compared with the SVM, it hasthe benefits of sparser model、the facility to utilize arbitrary kernel functions、more accurate、strong robusticity、 intensity generalization ability， and so on. RVM algorithm has beendeveloped rapidly in the fields of application, and proved better in pattern classification、faultdiagnosis、intelligent forecast、information processing of voice and image, etc. Whereas，tosolve the questions of multi-mode pattern recognition，RVM algorithm still suffer from theproblems of looking after both sides of accuracy and real-time because of the complexity ofcomputational process.
According to the insufficiency of RVM, the structure and key steps of the algorithm,including kernel functions selection, classifier construct, and control parameter setting, aredeeply investigated in this paper and improved on “one against one” classifier which has beenthe highest accuracy and broad application are proposed to improve. The improvedclassification method proposed is utilized successfully to reduce the time of classification andincrease in real-time without cutting down the accuracy. In addition, improved algorithm inthis paper is applied in face recognition&automobile engine fault diagnosis, and behavedwell.
Firstly, the research situation and the fundamental theory about RVM is detail discussedin this paper, and the solved key problem of RVM is presented. In order to improveclassification speed of multiclass pattern recognition based on relevance vector machinelearning algorithm， investigated the method of relevance vector machine algorithm inmulti-mode classification, and found that the comparison too many times was the main reasonfor large amount of calculation. Proposed a new waythat eliminated the most dissimilar classin each round of comparison. Comparison times were reduced step by step per cycle, and theclassification was more, the decrease in the total calculation amount was more obvious. Thevalidity of this method has been proved by some simulated examples, and the experimentalresults of data classification show that compared with traditional classifier, the training timesand the recognition times of the novel method are greatly reduced under the premise of hardly influence classification accuracy, the algorithm running speed is improved obviously.
Secondly, to solve the problems existed in face recognition, such as lack of accuracy、real-time and stability，a new face recognition approach based on improved relevance vectormachine is presented in this paper. Firstly, the wavelet transform is applied to preprocess faceimage to reduce the impact of expression change. Then, in order to extract key features of theprocessed face image, use the principal component analysis (PCA) method. Finally, the RVMclassification model is adopted for identifying. In comparison with the support vectormachine(SVM) method, the RVM approach performs well and can obtain more satisfactoryresults in terms of recognition rates、real-time and reliability.
Thirdly, the face images recognition accuracy will be obvious decline when the objectscontain more noise. At present, face recognition technology to solve this problem is no betterway. In this paper, a new method of face recognition based on relevance vector machine waspresented. After the wavelet decomposition and PCA transform, relevance vectors fromsample training constituted a "hyperplane" as the differences in the classification of thesamples by machine learning algorithm and used the improved “one against one” method toachieve multi-class pattern recognition. Compared with the former method, a large number ofsimulation results show that the new method used in noisy objects being recognized is notsensitive to image noise, with a more accurate and strong robusticity. In addition, photo light、angle change、occlusion、low resolution ratio and so on, also be discussed and analysed inexperiment with new method in this paper.
Finally, application of RVM in the automobile engine fault diagnosis is investigated.From the study we know that the parameter of penalty factor and kernel paraeter play a veryimportant roal on the diagnosis model, so the Particle Swarm Optimization(PSO) is used tooptimize the parameters, this algorithms practical applied to automotive engine fault diagnosis.Considering the problem of the variation of characteristic parameters are followed by enginerotational speed, puts forward a adaptive fitting of super-parameters on incremental learning.To the problem of engine misfire, mapping relations established between gas volume fractionand the cause of the misfire, used normalized data with different gears in machine training,ajusted super-parameters by curve fitting, and the trained RVM model applied in faultclassification and diagnosis.The simulation experiments shows the results of new method isnot only accurate and reliable but also resolve the problem of dynamic detection with variable speed in traditional methods.
In conclusion, a simple summary is made and some research aspects are presented in thefuture.
引文
[1] Tipping M E. Sparse Bayesian learning and the relevance vector machine[J]. machinelearning research,2001:211-244.
[2] ROBERT C, CASELLA G. Monte Carlo statistical methods[M]. Springer Verlag,2004.
[3] CORTES C, VAPNIK V. Support-vector networks [J].Machine learning,1995,20(3)：273-297.
[4] SEBALD D, BUCKLEW J. Support vector machine techniques for nonlinearequalization [J]. IEEE Transactions on Signal Processing,2000,48(11):3217-3226.
[5] Vapnik V.The nature of statistical learning theory[M]. New York: Springer,1998.
[6] Vapnic.统计学习理论的本质[M].张学工,译.北京:清华大学出版社,2000.
[7]唐发明.基于统计学习理论的支持向量机算法研究[D].武汉：华中科技大学,2005：67-81.
[8] Ioannis Psorakis, Theodoros Damoulas, Mark A, etc. Multiclass relevance vectormachines: sparsity and accuracy[C]//IEEE Transactions on neural networks,2010,21(10):1588-1598.
[9]赵春晖,张燚.相关向量机分类方法的研究进展与分析[J].智能系统学报,2012,7(4):1-8.
[10] ChellappaR,etal. Human and Machine Recognition of Faces: A Survey[J]. proceedingsof the IEEE,1995,83(5):705-740.
[11]艾英山，张德贤.人脸识别方法的综述与展望[J].计算机与数字工程,2005,33(10):24-27.
[12] Bottou L., Cortes C., Denker J.S. etal. Comparison of Classifier Methods: A Case Studyin Handwriting Digit Recognition[C]//International Conference on Pattern Recognition,IEEE Computer Society Press,1994:77~87.
[13]刘遵雄，张德运.基于相关向量机的电力负荷中期预测[J].西安交通大学学报.2004,38(10):1005-1008.
[14]陶新民，徐晶，杜宝祥，等.基于相空间RVM的轴承故障检测方法[J].振动与冲击,2007,25(12):1-7.
[15]马笑潇，黄席褪，柴毅.基于SVM的二叉树多类分类算法及其在故障诊断中的应用[J].控制与决策,2003,18(3):272-276.
[16]仕玉治，彭勇，周惠成.基于相关向量机的中长期径流预报模型研究[J].大连理工大学学报,2012,52(1):79-84.
[17]韩敏，孙磊磊，洪晓军，等.基于自回归模型和关联向量机的癫痫脑电信号自动分类[J].中国生物医学工程学报,2011,30(6)：864-870.
[18]杨成福，章毅.相关向量机及在说话人识别应用中的研究[J].电子科技大学学报,2010,39(2):311-315.
[19]张昱，谢小鹏.基于遗传相关向量机的图像分类技术[J].计算机仿真，2011，28(5):283-286.
[20]殷岳萌，冯燕，刘萌萌.基于独立分量分析和相关向量机的高光谱数据分类[J].现代电子技术,2010,13：123-126.
[21] JAAKKOLA T S, JORDAN M I. Bayesian parameter estimation through variationalmethods [J]. Statistics and Computing,2000,10(1):25-37.
[22] BISHOP C, TIPPING M. Variational relevance vector machines[C]//Proceedings of the16th Conference on Uncertainty in Artificial Intelligence,2000:46-53.
[23] M.E. Tipping. Sparse kernel principal component analysis [J].2001, NIPS.
[24] TIPPING M, FAUL A. Fast marginal likelihood maximization for sparse Bayesianmodels [C].(Citeseer),2003.
[25] Tipping.M.2004. Bayesian inference:An introduction to Principles and Practice inmachine learning. Advanced Lectures on Machine Learning,vol.3176:41-62.
[26] Thayananthan A. Template-based Pose Estimation and Tracking of3D Hand Motion[D]. Department of Engineering, University of Cambridge, September2005.
[27] Carl Edward Rasmussen,Christopher K.I.Williams. Gaussian Processes for MachineLearning. The MIT Press,2006,l:211-244.
[28] T. Damoulas, M. Girolami. Probabilistic multi-class multi-kernel learning: on proteinfold recognition and remote homology detection. Bioinformatics.2008,24(10):1264-1270.
[29]杨树仁，沈洪远.基于相关向量机的机器学习算法研究与应用[J].计算技术与自动化,2010,29(1):43-47.
[30]秦锋，杨波，程泽凯.分类器性能评价标准研究[J].计算机技术与发展，2006，16(17):85-88.
[31]邵信光，杨慧中，陈刚.基于粒子群优化算法的支持向量机参数选择及其应用[J].控制理论与应用，2006，23(5)：740-743.
[32]苑进.贝叶斯学习框架下非线性制造过程建模及多目标优化关键技术研究[D].上海:上海大学,2008.
[33]董争.基于相关向量机的大规模分类问题的研究[D].广州:华南理工大学,2009.
[34]李刚，王贵龙，薛惠锋. RVM核参数的遗传算法优化方法[J].控制工程，2010,17(3)：335-337.
[35] Smits G F, Jordan E M. Improved SVM Regression Using Mixtures[C]//Proc. of theInternational Joint Conference on Neural Networks. IEEE Press,2002:2785-2790.
[36]吴冰.相关向量回归元建模关键技术及其应用研究[D].长沙：国防科学技术大学，2011.
[37]杨柳，张磊，张少勋，等.单核和多核相关向量机的比较研究[J].计算机工程,2010,36(12):195-197.
[38] N.A.Syed, N.Liu, K.Sung. Incremental learnning with support vector machines.Workshop on support vector machines at the international joint conference on artificialintelligence(IJCAI-99), Stockholm,Sweden,1999.1104-1105.
[39] M.Martin. On-line support vector machines for function approximation.(Tech.ReP.LSI-02-11-R). Catalunya,Spain:Software department, Universitat Politecnica deCatalunya,2002.
[40]安金龙，王正欧.一种适合于增量学习的支持向量机的快速循环算法.计算机应用,2003,23(10):12-14,17.
[41] Nikolay Nikolaev, Peter Tino. Sequential Relevance Vector Machine Learning fromTime Series[C].2005,2:1308-1313.
[42]刘新旺.基于支持向量机的特征增量学习算法研究[D].长沙：国防科学技术大学，2008.
[43] Tom M. Mitchell.机器学习[M].曾华军，张银奎，译.北京：清华大学出版社，2003.
[44]边肇祺，张学工.模式识别(第2版)[M].北京:清华大学出版社,2000,1.
[45]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.
[46]辛宪会.支持向量机理论、算法与实现[D].郑州：中国人民解放军信息工程大学，2005.
[47]杨国鹏，周欣，余旭初.稀疏贝叶斯模型与相关向量机学习研究[J].计算机科学,2010,37(7)：225-228.
[48]宋晖，薛匀，张良均.基于SVM分类问题的核函数选择仿真研究[J].计算机与现代化，2011(8):133-136.
[49]肖健华.智能模式识别方法[M].华南理工大学出版社，2006,8.
[50]尹义龙，宁新宝，张晓梅.自动指纹识别技术的发展与应用[J].南京大学学报（自然科学），2002，38(1):29-35.
[51]张禹，马驷良，韩笑,等.车牌识别中的图像提取及分割算法[J].吉林大学学报（理学版）,2006,44(3):406-410.
[52]李中才，刘刚.矿井安全指标的灰色关联评价模型及应用[J].矿业工程,2006,4(4)：57-59.
[53]郭显娥，武伟，刘春贵，等.多类SVM分类算法的研究[J].山西大同大学学报（自然科学版）,2010,26(3):6-8.
[54] Thayananthan A. Relevance Vector Machine based Mixture of Experts [R].Departmentof Engineering, University of Cam-bridge,2005.
[55] Burges C J C. A tutorial on Support Vector Machines for Pattern Recognition[J].Knowledge Discovery and Data Mining,1998,2(2):121-167.
[56] Platt J, Cristianini N, Shawe-Taylor J. Large margin DAGs for multiclass classification.Leen T K, Müller K R. Advances in Neural Information Processing Systems12[C]. S ASolla: The MIT Press,2000:547-553.
[57] Schwenker F. Hierarchical support vector machines for multi-class patternrecognition[C]//Fourth International conference on Knowledge-Based IntelligentEngineering Systems&Allied Technologies. Brighton, UK,2000.
[58]李红莲,焦瑞莉,范京.支持向量机多类分类方法的精度分析[J].北京机械工业学院院报,2008,20(2):32-35.
[59] Weston J, Watkins C. Support vector machines for multi-class pattern recognition[D].In Proceedings of7th European Symposium on Artificial Neural Networks,1999:219-224.
[60]唐发明，王仲东，陈绵云.支持向量机多类分类算法研究[J].控制与决策,2005,20(7):746-754.
[61] Takahashi F, Abe S. Decision-Tree-Based Multiclass Support Vector Machines[C]//Proc of the9th Int Conf on Neural Information Processing. Singapore,2002(3):1418-1422.
[62]范柏超，王建宇，薄煜明.结合特征选择的二叉树SVM多分类算法[J].计算机工程与设计，2010,31(12):2823-2825.
[63]连可，黄建国，王厚军，等.一种基于遗传算法的SVM决策树多分类策略研究[J].2008,36(8):1502-1507.
[64]李雅琴，董才林，陈增照.一种新的基于SVM的多层分类方法[J].计算机工程与科学,2007,29(4):108-110.
[65]王艳，陈欢欢，沈毅.有向无环图的多类支持向量机分类算法[J].电机与控制学报，2011：85-89.
[66]李昆仑.一种基于有向无环图的多类SVM分类器[J].模式识别与人工智能，2003，16(2):164-168.
[67]刘志刚，李德仁，秦前清.支持向量机在多类分类问题中的推广[J].计算机工程与应用,2004,7:10-13.
[68] J. Ruiz-del-Solar, P. Navarrete. Eigenspace-based face recognition: a comparative studyof different approaches, IEEE Transactions on Systems, Man and Cybernetics, Part C,2005,35(3), pp:315-325.
[69]刘鹭，董立文，姜鹏，等.人脸识别技术的研究与应用[J].科技信息,2010,14:492.
[70] G. Guo, S. Z. Li and K. Chan. Face Recognition by Support Vector Machines[C]//Proc.of the4th Int. Conf. on Auto. Face and Gesture Recog., Grenoble,2000:196-201.
[71]杨洁，冯力刚，蒋加伏.基于小波包和支持向量机的人脸识别[J].计算机仿真,2004,21(9):131-133.
[72] A M Martinez，A C Kak. PCA versus LDA [J]. IEEE Transactions Pattern Analysis andMachine Intelligence,2001,23(2):228-233.
[73] KOTSIAI, ZAFEIRIOUS, PATASI. A novel discriminant non-negativematrixfactorization algorithm with applications to facial image characterization problems [J].IEEE Trans on Information Forensics and Security,2007,2(3):588-595.
[74]艾英山,张德贤.人脸识别方法的综述与展望[J].计算机与数字工程,2005,33(10):24-27.
[75] ChellappaR,etal. Human and Machine Recognition of Faces: A Survey[J]. proceedingsof the IEEE,1995,83(5):705-740.
[76] Pu Xiaorong， Zheng Ziming， Zhou Wei. An Improved Subspace Method forRecognizing Imprecisely Fragmentary Faces， Journal of University of ElectronicScience and Technology of China(in Chinese),2006,35(2):208-210.
[77]姚丽君.人脸识别技术的发展与应用[J].科技创新导报,2010,(21):29
[78] Lee HC, Gaensslen RE. Advances in Fingerpinrt Technology [M]. New York: Elsevier,1991.
[79]李华胜，杨桦，袁保宗.人脸识别系统中的特征提取[J].北方交通大学学报,2001,25(2):18-21.
[80] M. Turk, A. Pentland. Eigenfaces for Recognition. J. Cogn. Neuroscience,1991,3(1):71-86.
[81] F. X. Song, D. Zhang, J. Z. Wang, H. Liu, Q. Tao. A Parameterized Direct LDA and ItsApplication to Face Recognition. Neurocomputing,2007,71(1-3):191-196.
[82]黄璞，陈才扣.基于二维图像矩阵的ICA人脸识别[J].计算机工程与设计,2009,30(24):5686-5688.
[83] D. S. Bolme. Elastic Bunch Graph Matching. Thesis for the Degree of Master ofScience in Computer Science, Colorado State University. Colorado, USA.2003.
[84] Nefian A V, Hayes M H. Face Detection and Recognition Using Hidden MarkovModels [J]. International Conference on Image Processing,2002,15(1):141-145.
[85]蔺广逢，范引娣，张媛.主成分分析与BP神经网络的人脸识别方法研究[J].现代电子技术,2007(2)：53-55.
[86] Yang L H, Bui T D, Suen C Y. An Application of Nonlinear Wavelet Approximation toFace Recognition. Pattern Recognition[C]. Proceedings16th International Conferenceon.2002,2(11-15):48-51.
[87]王剑平，张捷.小波变换在数字图像处理中的应用[J].现代电子技术,2011,34(1):91-94.
[88]郑德忠，崔法毅.基于小波变换与小域特征模糊融合的人脸识别[J].光学技术,2008,34(6):841-846.
[89]陶晓燕，刘振霞，王元一.基于小波子带的PCA人脸识别方法研究[J].空军工程大学学报(自然科学版),2004,5(3):65-67.
[90]孙琳，秦文华，吴冬梅.基于PCA和核FDA的人脸识别研究[J].通信技术,2011,44(4):19-20.
[91]舒双宝，罗家融，徐从东，等.一种基于支持向量机的人脸识别新方法[J].计算机仿真,2011,28(2):280-283.
[92]黄玉程，胡国清，吴雄英，等.人脸识别系统中图像噪声去除方法研究[J].微计算机信息(嵌入式与SOC),2005,21(12):187-188.
[93]宋寅卯，李晓娟，刘磊.图像噪声滤波的研究方法及进展[J].电脑开发与应用,2010,23(4):74-76.
[94]王民，文义铃.常用图像去噪算法的比较与研究[J].西安建筑科技大学学报(自然科学版),2010,42(6):895-898.
[95] Coyle E J,Gabbouj M,Lin J H.1991, From Median Filters to Optimal Stack Filtering [J].In: IEEE Internat. Symp. Circuits Systems,1991,1:9-12.
[96] Ko S J, Lee Y H.Center Weighted Median Filter and their Application to ImageEnhancement[J]. IEEE Trans. circ. Syst,1991,38:984-933.
[97] Lin H M,Willson A N.Median Filters with Adaptive Length [J]. IEEE Transactions onCircuits and Systems,1988,35(6):675-690.
[98]陈乃金，周鸣争，潘冬冬.一种新的维纳滤波图像去高斯噪声算法[J].计算机系统应用,2010,19(3):111-114.
[99]赵瑞珍.小波理论及其在图像信号处理中的算法研究[D].西安:西安电子科技大学,2002.
[100] Xu Y. Wavelet Transform Domain Filters: a Spatially Selective Noise FiltrationTechnique[J]. IEEE Trans on IP,1994,3(6):217-237.
[101] Mallat S, Hwang W L. Singularity Detection and Processing with Wavelets [J]. IEEETrans on IT,1992,38(2):612-643.
[102]姜珊，王跃存.基于中值滤波和形态学的去噪算法[J].仪器仪表用户,2007,14(4):106-107.
[103] Ora Y, Wilamowski B M. Analog Implementation of Pulse-coupled Neural Networks[J]. IEEE Trans. Neural Networks,1999,10(3):539-543.
[104] Johnson J L, Padgett M L. PCNN Model and Applications [J]. IEEE Transaction onNeural Networks,1999,10(3):480-498.
[105]方莉，张萍.经典图像去噪算法研究综述[J].工业控制计算机,2010,23(11):73-74.
[106]李树涛，王耀南，龚理专.多聚焦图像融合中最佳小波分解层数的选取[J].系统工程与电子技术,2002,24(6):45-48.
[107]李占述，叶海霞，徐伯庆.一种数字图像随机噪声的估计及利用MATLAB的实现[J].装备制造技术,2009,6:1-2.
[108]刘宇，王国裕. CMOS图像传感器固定模式噪声抑制新技术[J].固体电子学研究与进展,2006,26(3):345-349.
[109]唐斌兵，王正明，汪雄良.一种含椒盐噪声图像去噪的新方法[J].系统工程,2008,26,(10):123-126.
[110]曹文庆，张柏荣，杨为民.具有泊松噪声分布的图像复原方法[J].云南天文台台刊,1994,2:36-48.
[111] X.H.Han，Y.W.Chen，Z.Nakao. An ICA-based Method for Poisson Noise Reduction[J].Lecture Notes in Artificial Intelligence,2003,2773:1449-1454.
[112]孟芳兵，吴雷.几种应用于人脸识别的光照预处理方法对比[J].武汉理工大学学报(信息与管理工程版),2010,32(4):542-546.
[113]陈熙霖，山世光，高文.多姿态人脸识别[J].中国图象图形学报,1999,4(10):818-824.
[114]章品正，赵洪玉，梁晓云，等.一种复杂背景中的人脸检测与验证方法[J].数据采集与处理,2004,19(1):10-15.
[115]邹利华.检测彩色图像中人头数的人脸分割算法[J].广州白云学院院报,2009,16(2):96-98.
[116]吴广.汽车故障诊断系统研究[D].长春:吉林大学,2009:1-5.
[117]张丽莉,储江伟,强添刚.现代汽车故障诊断方法及其应用研究[J].机械研究与应用,2008,21(1):8-16．
[118] RIZZONI G, MIN P.S. Detection of Sensor Failures in Automotive Engines[C].Vehicular Technology. IEEE Transactions, USA,1991,40(2):487-500.
[119] YONG W K, GRORGIO R. Automotive Engine Diagnosis and Control Via NonlinearEstimation Control System[J]. Microprocessors and Microsystems,1998,8(1):3-7.
[120] KIHOON CHOI, SINGH S, KODALI A, et al. Novel Classifier Fusion Approaches forFault Diagnosis[C]. Instrumentation and Measurement. IEEE Transactions, USA,2009,58(3):602-611.
[121] YONG W K, GRORGIO R. Automotive Engine Diagnosis and Control Via NonlinearEstimation Control System[J]. Microprocessors and Microsystems,1998,8(1):3-7.
[122]张博.基于CAN总线信息的汽车故障分析及预报系统的研究[D].哈尔滨:哈尔滨理工大学,2009:17-19.
[123]马忠梅,祝烈煌,李善平,叶楠. ARM&Linux嵌入式系统教程(第二版)[M].北京:北京航空航天大学出版社,2008:199-205.
[124]陈立新，刘福建，何玉灵.汽车发动机智能故障诊断方法综述[J].仪器仪表与分析监测,2008,(2):1-3.
[125]吴今培.智能故障诊断与专家系统[M].北京:科学出版社,1999.
[126]周志英. FUZZY数学在汽车电喷发动机故障诊断中的应用[J].湖南工程学院学报,2006,16(1):43-45.
[127]施云.模糊故障树在汽车发动机故障诊断中的应用[J].桂林电子科技大学学报,2008,28(3):222-223.
[128]朱惠莲.基于神经网络的汽车发动机故障诊断系统[D].杨凌：西北农林科技大学,2007.
[129]付金.基于GA优化BP神经网络的汽车发动机失火故障诊断研究[D].沈阳:东北大学,2009.
[130] Lu Di, Dou Wen-juan. Fault diagnosis of engine misfire based on genetic optimizedsupport vector machine[C]//The6thInternational Forum on Strategic Technology,Harbin, China,2011:1223-1225.
[131]窦文娟.基于ARM9的车载故障诊断系统设计[D].哈尔滨:哈尔滨理工大学,2012.
[132]徐元强,施红星,苏建城.汽车发动机检测诊断技术[M].北京:电子工业出版社,2006:10-45.
[133]陈煜.车辆发动机检测评估诊断系统研究[D].天津:天津大学,2009:1-4.
[134]肖云魁.汽车故障诊断学[M].北京:北京理工大学出版社,2001.
[135]张忠伟，沈憬虹，曹友春.汽车尾气成份及故障诊断[J].漯河职业技术学院学报,2008(2):26-28.
[136]赵远,张宇.光电信号检测原理与技术[M].北京:机械工业出版社,2005:33-35.
[137]王凯,张永祥,李军.基于支持向量机的齿轮故障诊断方法研究[J].国外电子测量技术,2009,28(5):49-51.
[138]夏天，王新晴，肖云魁，等.基于离散粒子群优化算法的汽车发动机故障特征选择[J].中国工程机械学报,2010,8(2):219-223.
[139]吴良海.基于粒子群优化相关向量机的网络入侵检测[J].微电子学与计算机,2010,27(5):181-184.
[140] J.RATSABY. Incremental learning with sample queries. IEEE Transactions on PatternAnalysis and Machine Intelligence,1998,20(8):883-888.
[141] L.RALAIVOLA AND F.D′ALCHé-BUC. Incremental support vector machine: a localapproach. In Proc.Int.Conf. on Artificial Neural Networks, ICANN’2001,Vienne,Autriche,2001.
[142]峦杰，唐常杰，黄晓冬，等.一种增量式支持向量机文本分类模型[J].计算机科学,2008,30(10):244-246.
[2] ROBERT C, CASELLA G. Monte Carlo statistical methods[M]. Springer Verlag,2004.
[3] CORTES C, VAPNIK V. Support-vector networks [J].Machine learning,1995,20(3)：273-297.
[4] SEBALD D, BUCKLEW J. Support vector machine techniques for nonlinearequalization [J]. IEEE Transactions on Signal Processing,2000,48(11):3217-3226.
[5] Vapnik V.The nature of statistical learning theory[M]. New York: Springer,1998.
[6] Vapnic.统计学习理论的本质[M].张学工,译.北京:清华大学出版社,2000.
[7]唐发明.基于统计学习理论的支持向量机算法研究[D].武汉：华中科技大学,2005：67-81.
[8] Ioannis Psorakis, Theodoros Damoulas, Mark A, etc. Multiclass relevance vectormachines: sparsity and accuracy[C]//IEEE Transactions on neural networks,2010,21(10):1588-1598.
[9]赵春晖,张燚.相关向量机分类方法的研究进展与分析[J].智能系统学报,2012,7(4):1-8.
[10] ChellappaR,etal. Human and Machine Recognition of Faces: A Survey[J]. proceedingsof the IEEE,1995,83(5):705-740.
[11]艾英山，张德贤.人脸识别方法的综述与展望[J].计算机与数字工程,2005,33(10):24-27.
[12] Bottou L., Cortes C., Denker J.S. etal. Comparison of Classifier Methods: A Case Studyin Handwriting Digit Recognition[C]//International Conference on Pattern Recognition,IEEE Computer Society Press,1994:77~87.
[13]刘遵雄，张德运.基于相关向量机的电力负荷中期预测[J].西安交通大学学报.2004,38(10):1005-1008.
[14]陶新民，徐晶，杜宝祥，等.基于相空间RVM的轴承故障检测方法[J].振动与冲击,2007,25(12):1-7.
[15]马笑潇，黄席褪，柴毅.基于SVM的二叉树多类分类算法及其在故障诊断中的应用[J].控制与决策,2003,18(3):272-276.
[16]仕玉治，彭勇，周惠成.基于相关向量机的中长期径流预报模型研究[J].大连理工大学学报,2012,52(1):79-84.
[17]韩敏，孙磊磊，洪晓军，等.基于自回归模型和关联向量机的癫痫脑电信号自动分类[J].中国生物医学工程学报,2011,30(6)：864-870.
[18]杨成福，章毅.相关向量机及在说话人识别应用中的研究[J].电子科技大学学报,2010,39(2):311-315.
[19]张昱，谢小鹏.基于遗传相关向量机的图像分类技术[J].计算机仿真，2011，28(5):283-286.
[20]殷岳萌，冯燕，刘萌萌.基于独立分量分析和相关向量机的高光谱数据分类[J].现代电子技术,2010,13：123-126.
[21] JAAKKOLA T S, JORDAN M I. Bayesian parameter estimation through variationalmethods [J]. Statistics and Computing,2000,10(1):25-37.
[22] BISHOP C, TIPPING M. Variational relevance vector machines[C]//Proceedings of the16th Conference on Uncertainty in Artificial Intelligence,2000:46-53.
[23] M.E. Tipping. Sparse kernel principal component analysis [J].2001, NIPS.
[24] TIPPING M, FAUL A. Fast marginal likelihood maximization for sparse Bayesianmodels [C].(Citeseer),2003.
[25] Tipping.M.2004. Bayesian inference:An introduction to Principles and Practice inmachine learning. Advanced Lectures on Machine Learning,vol.3176:41-62.
[26] Thayananthan A. Template-based Pose Estimation and Tracking of3D Hand Motion[D]. Department of Engineering, University of Cambridge, September2005.
[27] Carl Edward Rasmussen,Christopher K.I.Williams. Gaussian Processes for MachineLearning. The MIT Press,2006,l:211-244.
[28] T. Damoulas, M. Girolami. Probabilistic multi-class multi-kernel learning: on proteinfold recognition and remote homology detection. Bioinformatics.2008,24(10):1264-1270.
[29]杨树仁，沈洪远.基于相关向量机的机器学习算法研究与应用[J].计算技术与自动化,2010,29(1):43-47.
[30]秦锋，杨波，程泽凯.分类器性能评价标准研究[J].计算机技术与发展，2006，16(17):85-88.
[31]邵信光，杨慧中，陈刚.基于粒子群优化算法的支持向量机参数选择及其应用[J].控制理论与应用，2006，23(5)：740-743.
[32]苑进.贝叶斯学习框架下非线性制造过程建模及多目标优化关键技术研究[D].上海:上海大学,2008.
[33]董争.基于相关向量机的大规模分类问题的研究[D].广州:华南理工大学,2009.
[34]李刚，王贵龙，薛惠锋. RVM核参数的遗传算法优化方法[J].控制工程，2010,17(3)：335-337.
[35] Smits G F, Jordan E M. Improved SVM Regression Using Mixtures[C]//Proc. of theInternational Joint Conference on Neural Networks. IEEE Press,2002:2785-2790.
[36]吴冰.相关向量回归元建模关键技术及其应用研究[D].长沙：国防科学技术大学，2011.
[37]杨柳，张磊，张少勋，等.单核和多核相关向量机的比较研究[J].计算机工程,2010,36(12):195-197.
[38] N.A.Syed, N.Liu, K.Sung. Incremental learnning with support vector machines.Workshop on support vector machines at the international joint conference on artificialintelligence(IJCAI-99), Stockholm,Sweden,1999.1104-1105.
[39] M.Martin. On-line support vector machines for function approximation.(Tech.ReP.LSI-02-11-R). Catalunya,Spain:Software department, Universitat Politecnica deCatalunya,2002.
[40]安金龙，王正欧.一种适合于增量学习的支持向量机的快速循环算法.计算机应用,2003,23(10):12-14,17.
[41] Nikolay Nikolaev, Peter Tino. Sequential Relevance Vector Machine Learning fromTime Series[C].2005,2:1308-1313.
[42]刘新旺.基于支持向量机的特征增量学习算法研究[D].长沙：国防科学技术大学，2008.
[43] Tom M. Mitchell.机器学习[M].曾华军，张银奎，译.北京：清华大学出版社，2003.
[44]边肇祺，张学工.模式识别(第2版)[M].北京:清华大学出版社,2000,1.
[45]张学工.关于统计学习理论与支持向量机[J].自动化学报,2000,26(1):32-42.
[46]辛宪会.支持向量机理论、算法与实现[D].郑州：中国人民解放军信息工程大学，2005.
[47]杨国鹏，周欣，余旭初.稀疏贝叶斯模型与相关向量机学习研究[J].计算机科学,2010,37(7)：225-228.
[48]宋晖，薛匀，张良均.基于SVM分类问题的核函数选择仿真研究[J].计算机与现代化，2011(8):133-136.
[49]肖健华.智能模式识别方法[M].华南理工大学出版社，2006,8.
[50]尹义龙，宁新宝，张晓梅.自动指纹识别技术的发展与应用[J].南京大学学报（自然科学），2002，38(1):29-35.
[51]张禹，马驷良，韩笑,等.车牌识别中的图像提取及分割算法[J].吉林大学学报（理学版）,2006,44(3):406-410.
[52]李中才，刘刚.矿井安全指标的灰色关联评价模型及应用[J].矿业工程,2006,4(4)：57-59.
[53]郭显娥，武伟，刘春贵，等.多类SVM分类算法的研究[J].山西大同大学学报（自然科学版）,2010,26(3):6-8.
[54] Thayananthan A. Relevance Vector Machine based Mixture of Experts [R].Departmentof Engineering, University of Cam-bridge,2005.
[55] Burges C J C. A tutorial on Support Vector Machines for Pattern Recognition[J].Knowledge Discovery and Data Mining,1998,2(2):121-167.
[56] Platt J, Cristianini N, Shawe-Taylor J. Large margin DAGs for multiclass classification.Leen T K, Müller K R. Advances in Neural Information Processing Systems12[C]. S ASolla: The MIT Press,2000:547-553.
[57] Schwenker F. Hierarchical support vector machines for multi-class patternrecognition[C]//Fourth International conference on Knowledge-Based IntelligentEngineering Systems&Allied Technologies. Brighton, UK,2000.
[58]李红莲,焦瑞莉,范京.支持向量机多类分类方法的精度分析[J].北京机械工业学院院报,2008,20(2):32-35.
[59] Weston J, Watkins C. Support vector machines for multi-class pattern recognition[D].In Proceedings of7th European Symposium on Artificial Neural Networks,1999:219-224.
[60]唐发明，王仲东，陈绵云.支持向量机多类分类算法研究[J].控制与决策,2005,20(7):746-754.
[61] Takahashi F, Abe S. Decision-Tree-Based Multiclass Support Vector Machines[C]//Proc of the9th Int Conf on Neural Information Processing. Singapore,2002(3):1418-1422.
[62]范柏超，王建宇，薄煜明.结合特征选择的二叉树SVM多分类算法[J].计算机工程与设计，2010,31(12):2823-2825.
[63]连可，黄建国，王厚军，等.一种基于遗传算法的SVM决策树多分类策略研究[J].2008,36(8):1502-1507.
[64]李雅琴，董才林，陈增照.一种新的基于SVM的多层分类方法[J].计算机工程与科学,2007,29(4):108-110.
[65]王艳，陈欢欢，沈毅.有向无环图的多类支持向量机分类算法[J].电机与控制学报，2011：85-89.
[66]李昆仑.一种基于有向无环图的多类SVM分类器[J].模式识别与人工智能，2003，16(2):164-168.
[67]刘志刚，李德仁，秦前清.支持向量机在多类分类问题中的推广[J].计算机工程与应用,2004,7:10-13.
[68] J. Ruiz-del-Solar, P. Navarrete. Eigenspace-based face recognition: a comparative studyof different approaches, IEEE Transactions on Systems, Man and Cybernetics, Part C,2005,35(3), pp:315-325.
[69]刘鹭，董立文，姜鹏，等.人脸识别技术的研究与应用[J].科技信息,2010,14:492.
[70] G. Guo, S. Z. Li and K. Chan. Face Recognition by Support Vector Machines[C]//Proc.of the4th Int. Conf. on Auto. Face and Gesture Recog., Grenoble,2000:196-201.
[71]杨洁，冯力刚，蒋加伏.基于小波包和支持向量机的人脸识别[J].计算机仿真,2004,21(9):131-133.
[72] A M Martinez，A C Kak. PCA versus LDA [J]. IEEE Transactions Pattern Analysis andMachine Intelligence,2001,23(2):228-233.
[73] KOTSIAI, ZAFEIRIOUS, PATASI. A novel discriminant non-negativematrixfactorization algorithm with applications to facial image characterization problems [J].IEEE Trans on Information Forensics and Security,2007,2(3):588-595.
[74]艾英山,张德贤.人脸识别方法的综述与展望[J].计算机与数字工程,2005,33(10):24-27.
[75] ChellappaR,etal. Human and Machine Recognition of Faces: A Survey[J]. proceedingsof the IEEE,1995,83(5):705-740.
[76] Pu Xiaorong， Zheng Ziming， Zhou Wei. An Improved Subspace Method forRecognizing Imprecisely Fragmentary Faces， Journal of University of ElectronicScience and Technology of China(in Chinese),2006,35(2):208-210.
[77]姚丽君.人脸识别技术的发展与应用[J].科技创新导报,2010,(21):29
[78] Lee HC, Gaensslen RE. Advances in Fingerpinrt Technology [M]. New York: Elsevier,1991.
[79]李华胜，杨桦，袁保宗.人脸识别系统中的特征提取[J].北方交通大学学报,2001,25(2):18-21.
[80] M. Turk, A. Pentland. Eigenfaces for Recognition. J. Cogn. Neuroscience,1991,3(1):71-86.
[81] F. X. Song, D. Zhang, J. Z. Wang, H. Liu, Q. Tao. A Parameterized Direct LDA and ItsApplication to Face Recognition. Neurocomputing,2007,71(1-3):191-196.
[82]黄璞，陈才扣.基于二维图像矩阵的ICA人脸识别[J].计算机工程与设计,2009,30(24):5686-5688.
[83] D. S. Bolme. Elastic Bunch Graph Matching. Thesis for the Degree of Master ofScience in Computer Science, Colorado State University. Colorado, USA.2003.
[84] Nefian A V, Hayes M H. Face Detection and Recognition Using Hidden MarkovModels [J]. International Conference on Image Processing,2002,15(1):141-145.
[85]蔺广逢，范引娣，张媛.主成分分析与BP神经网络的人脸识别方法研究[J].现代电子技术,2007(2)：53-55.
[86] Yang L H, Bui T D, Suen C Y. An Application of Nonlinear Wavelet Approximation toFace Recognition. Pattern Recognition[C]. Proceedings16th International Conferenceon.2002,2(11-15):48-51.
[87]王剑平，张捷.小波变换在数字图像处理中的应用[J].现代电子技术,2011,34(1):91-94.
[88]郑德忠，崔法毅.基于小波变换与小域特征模糊融合的人脸识别[J].光学技术,2008,34(6):841-846.
[89]陶晓燕，刘振霞，王元一.基于小波子带的PCA人脸识别方法研究[J].空军工程大学学报(自然科学版),2004,5(3):65-67.
[90]孙琳，秦文华，吴冬梅.基于PCA和核FDA的人脸识别研究[J].通信技术,2011,44(4):19-20.
[91]舒双宝，罗家融，徐从东，等.一种基于支持向量机的人脸识别新方法[J].计算机仿真,2011,28(2):280-283.
[92]黄玉程，胡国清，吴雄英，等.人脸识别系统中图像噪声去除方法研究[J].微计算机信息(嵌入式与SOC),2005,21(12):187-188.
[93]宋寅卯，李晓娟，刘磊.图像噪声滤波的研究方法及进展[J].电脑开发与应用,2010,23(4):74-76.
[94]王民，文义铃.常用图像去噪算法的比较与研究[J].西安建筑科技大学学报(自然科学版),2010,42(6):895-898.
[95] Coyle E J,Gabbouj M,Lin J H.1991, From Median Filters to Optimal Stack Filtering [J].In: IEEE Internat. Symp. Circuits Systems,1991,1:9-12.
[96] Ko S J, Lee Y H.Center Weighted Median Filter and their Application to ImageEnhancement[J]. IEEE Trans. circ. Syst,1991,38:984-933.
[97] Lin H M,Willson A N.Median Filters with Adaptive Length [J]. IEEE Transactions onCircuits and Systems,1988,35(6):675-690.
[98]陈乃金，周鸣争，潘冬冬.一种新的维纳滤波图像去高斯噪声算法[J].计算机系统应用,2010,19(3):111-114.
[99]赵瑞珍.小波理论及其在图像信号处理中的算法研究[D].西安:西安电子科技大学,2002.
[100] Xu Y. Wavelet Transform Domain Filters: a Spatially Selective Noise FiltrationTechnique[J]. IEEE Trans on IP,1994,3(6):217-237.
[101] Mallat S, Hwang W L. Singularity Detection and Processing with Wavelets [J]. IEEETrans on IT,1992,38(2):612-643.
[102]姜珊，王跃存.基于中值滤波和形态学的去噪算法[J].仪器仪表用户,2007,14(4):106-107.
[103] Ora Y, Wilamowski B M. Analog Implementation of Pulse-coupled Neural Networks[J]. IEEE Trans. Neural Networks,1999,10(3):539-543.
[104] Johnson J L, Padgett M L. PCNN Model and Applications [J]. IEEE Transaction onNeural Networks,1999,10(3):480-498.
[105]方莉，张萍.经典图像去噪算法研究综述[J].工业控制计算机,2010,23(11):73-74.
[106]李树涛，王耀南，龚理专.多聚焦图像融合中最佳小波分解层数的选取[J].系统工程与电子技术,2002,24(6):45-48.
[107]李占述，叶海霞，徐伯庆.一种数字图像随机噪声的估计及利用MATLAB的实现[J].装备制造技术,2009,6:1-2.
[108]刘宇，王国裕. CMOS图像传感器固定模式噪声抑制新技术[J].固体电子学研究与进展,2006,26(3):345-349.
[109]唐斌兵，王正明，汪雄良.一种含椒盐噪声图像去噪的新方法[J].系统工程,2008,26,(10):123-126.
[110]曹文庆，张柏荣，杨为民.具有泊松噪声分布的图像复原方法[J].云南天文台台刊,1994,2:36-48.
[111] X.H.Han，Y.W.Chen，Z.Nakao. An ICA-based Method for Poisson Noise Reduction[J].Lecture Notes in Artificial Intelligence,2003,2773:1449-1454.
[112]孟芳兵，吴雷.几种应用于人脸识别的光照预处理方法对比[J].武汉理工大学学报(信息与管理工程版),2010,32(4):542-546.
[113]陈熙霖，山世光，高文.多姿态人脸识别[J].中国图象图形学报,1999,4(10):818-824.
[114]章品正，赵洪玉，梁晓云，等.一种复杂背景中的人脸检测与验证方法[J].数据采集与处理,2004,19(1):10-15.
[115]邹利华.检测彩色图像中人头数的人脸分割算法[J].广州白云学院院报,2009,16(2):96-98.
[116]吴广.汽车故障诊断系统研究[D].长春:吉林大学,2009:1-5.
[117]张丽莉,储江伟,强添刚.现代汽车故障诊断方法及其应用研究[J].机械研究与应用,2008,21(1):8-16．
[118] RIZZONI G, MIN P.S. Detection of Sensor Failures in Automotive Engines[C].Vehicular Technology. IEEE Transactions, USA,1991,40(2):487-500.
[119] YONG W K, GRORGIO R. Automotive Engine Diagnosis and Control Via NonlinearEstimation Control System[J]. Microprocessors and Microsystems,1998,8(1):3-7.
[120] KIHOON CHOI, SINGH S, KODALI A, et al. Novel Classifier Fusion Approaches forFault Diagnosis[C]. Instrumentation and Measurement. IEEE Transactions, USA,2009,58(3):602-611.
[121] YONG W K, GRORGIO R. Automotive Engine Diagnosis and Control Via NonlinearEstimation Control System[J]. Microprocessors and Microsystems,1998,8(1):3-7.
[122]张博.基于CAN总线信息的汽车故障分析及预报系统的研究[D].哈尔滨:哈尔滨理工大学,2009:17-19.
[123]马忠梅,祝烈煌,李善平,叶楠. ARM&Linux嵌入式系统教程(第二版)[M].北京:北京航空航天大学出版社,2008:199-205.
[124]陈立新，刘福建，何玉灵.汽车发动机智能故障诊断方法综述[J].仪器仪表与分析监测,2008,(2):1-3.
[125]吴今培.智能故障诊断与专家系统[M].北京:科学出版社,1999.
[126]周志英. FUZZY数学在汽车电喷发动机故障诊断中的应用[J].湖南工程学院学报,2006,16(1):43-45.
[127]施云.模糊故障树在汽车发动机故障诊断中的应用[J].桂林电子科技大学学报,2008,28(3):222-223.
[128]朱惠莲.基于神经网络的汽车发动机故障诊断系统[D].杨凌：西北农林科技大学,2007.
[129]付金.基于GA优化BP神经网络的汽车发动机失火故障诊断研究[D].沈阳:东北大学,2009.
[130] Lu Di, Dou Wen-juan. Fault diagnosis of engine misfire based on genetic optimizedsupport vector machine[C]//The6thInternational Forum on Strategic Technology,Harbin, China,2011:1223-1225.
[131]窦文娟.基于ARM9的车载故障诊断系统设计[D].哈尔滨:哈尔滨理工大学,2012.
[132]徐元强,施红星,苏建城.汽车发动机检测诊断技术[M].北京:电子工业出版社,2006:10-45.
[133]陈煜.车辆发动机检测评估诊断系统研究[D].天津:天津大学,2009:1-4.
[134]肖云魁.汽车故障诊断学[M].北京:北京理工大学出版社,2001.
[135]张忠伟，沈憬虹，曹友春.汽车尾气成份及故障诊断[J].漯河职业技术学院学报,2008(2):26-28.
[136]赵远,张宇.光电信号检测原理与技术[M].北京:机械工业出版社,2005:33-35.
[137]王凯,张永祥,李军.基于支持向量机的齿轮故障诊断方法研究[J].国外电子测量技术,2009,28(5):49-51.
[138]夏天，王新晴，肖云魁，等.基于离散粒子群优化算法的汽车发动机故障特征选择[J].中国工程机械学报,2010,8(2):219-223.
[139]吴良海.基于粒子群优化相关向量机的网络入侵检测[J].微电子学与计算机,2010,27(5):181-184.
[140] J.RATSABY. Incremental learning with sample queries. IEEE Transactions on PatternAnalysis and Machine Intelligence,1998,20(8):883-888.
[141] L.RALAIVOLA AND F.D′ALCHé-BUC. Incremental support vector machine: a localapproach. In Proc.Int.Conf. on Artificial Neural Networks, ICANN’2001,Vienne,Autriche,2001.
[142]峦杰，唐常杰，黄晓冬，等.一种增量式支持向量机文本分类模型[J].计算机科学,2008,30(10):244-246.
© 2004-2018 中国地质图书馆版权所有 京ICP备05064691号 京公网安备11010802017129号 地址：北京市海淀区学院路29号 邮编：100083 电话：办公室：(+86 10)66554848；文献借阅、咨询服务、科技查新：66554700 |