Based on the least-square minimization a computationally efficient learning algorithm for the Principal Component Analysis(PCA) is derived. The dual learning rate parameters are adaptively introduced to make the propo...Based on the least-square minimization a computationally efficient learning algorithm for the Principal Component Analysis(PCA) is derived. The dual learning rate parameters are adaptively introduced to make the proposed algorithm providing the capability of the fast convergence and high accuracy for extracting all the principal components. It is shown that all the information needed for PCA can be completely represented by the unnormalized weight vector which is updated based only on the corresponding neuron input-output product. The convergence performance of the proposed algorithm is briefly analyzed.The relation between Oja’s rule and the least squares learning rule is also established. Finally, a simulation example is given to illustrate the effectiveness of this algorithm for PCA.展开更多
Discriminating internal layers by radio echo sounding is important in analyzing the thickness and ice deposits in the Antarctic ice sheet.The signal processing method of synthesis aperture radar(SAR)has been widely us...Discriminating internal layers by radio echo sounding is important in analyzing the thickness and ice deposits in the Antarctic ice sheet.The signal processing method of synthesis aperture radar(SAR)has been widely used for improving the signal to noise ratio(SNR)and discriminating internal layers by radio echo sounding data of ice sheets.This method is not efficient when we use edge detection operators to obtain accurate information of the layers,especially the ice-bed interface.This paper presents a new image processing method via a combined robust principal component analysis-total variation(RPCA-TV)approach for discriminating internal layers of ice sheets by radio echo sounding data.The RPCA-based method is adopted to project the high-dimensional observations to low-dimensional subspace structure to accelerate the operation of the TV-based method,which is used to discriminate the internal layers.The efficiency of the presented method has been tested on simulation data and the dataset of the Institute of Electronics,Chinese Academy of Sciences,collected during CHINARE 28.The results show that the new method is more efficient than the previous method in discriminating internal layers of ice sheets by radio echo sounding data.展开更多
Multiblock kernel principal component analysis (MBKPCA) has been proposed to isolate the faults and avoid the high computation cost. However, MBKPCA is not available for dynamic processes. To solve this problem, recur...Multiblock kernel principal component analysis (MBKPCA) has been proposed to isolate the faults and avoid the high computation cost. However, MBKPCA is not available for dynamic processes. To solve this problem, recursive MBKPCA is proposed for monitoring large scale processes. In this paper, we present a new recursive MBKPCA (RMBKPCA) algorithm, where the adaptive technique is adopted for dynamic characteristics. The proposed algorithm reduces the high computation cost, and is suitable for online model updating in the feature space. The proposed algorithm was applied to an industrial process for adaptive monitoring and found to efficiently capture the time-varying and nonlinear relationship in the process variables.展开更多
The paper analyzes the problem of blind source separation （BSS） based on the nonlinear principal component analysis （NPCA） criterion. An adaptive strong tracking filter （STF） based algorithm was developed, which...The paper analyzes the problem of blind source separation （BSS） based on the nonlinear principal component analysis （NPCA） criterion. An adaptive strong tracking filter （STF） based algorithm was developed, which is immune to system model mismatches. Simulations demonstrate that the algorithm converges quickly and has satisfactory steady-state accuracy. The Kalman filtering algorithm and the recursive leastsquares type algorithm are shown to be special cases of the STF algorithm. Since the forgetting factor is adaptively updated by adjustment of the Kalman gain, the STF scheme provides more powerful tracking capability than the Kalman filtering algorithm and recursive least-squares algorithm.展开更多
近年来,鲁棒主成分分析法(Robust Principal Component Analysis,RPCA)被广泛应用到运动目标检测中,但该类方法未能有效利用运动目标的时空连续性先验,容易将动态背景误判为运动目标,且背景恢复精度不高.为此提出一种基于全变分-核回归...近年来,鲁棒主成分分析法(Robust Principal Component Analysis,RPCA)被广泛应用到运动目标检测中,但该类方法未能有效利用运动目标的时空连续性先验,容易将动态背景误判为运动目标,且背景恢复精度不高.为此提出一种基于全变分-核回归的RPCA运动目标检测方法.该方法以RPCA为基础,利用3维全变分模型增强前景的时空连续性,去除动态背景干扰,得到清晰完整的前景.同时,利用基于扩散张量的核回归对背景的时空相关性建模,去除噪声干扰,从而精确恢复背景.在多组公开数据集上的实验结果表明,该方法在动态背景、光照变化等复杂场景中能够较为精确地检测出运动目标和恢复背景.展开更多
提出一种改进的基于数据块更新的递归主元分析(recursive principal component analysis,RPCA)方法,对具有慢时变和多变量等特性的某型舰空导弹武器雷达发射机工作过程进行自适应监测。该方法在协方差矩阵的特征值分解中引入低秩奇异值...提出一种改进的基于数据块更新的递归主元分析(recursive principal component analysis,RPCA)方法,对具有慢时变和多变量等特性的某型舰空导弹武器雷达发射机工作过程进行自适应监测。该方法在协方差矩阵的特征值分解中引入低秩奇异值分解递归方法,实现负荷矩阵和特征值矩阵的递归计算;制定了均值、方差的更新策略;给出一种基于指数加权的控制限递归算法以提高RPCA的健壮性。实验证明该方法能自适应地跟踪过程时变并实时监测故障,同时有效地降低误警率。展开更多
鲁棒主成分分析(Robust principal component analysis,RPCA)模型中秩函数和L0范数的求解是非确定性多项式(Nondeterministic polynominal,NP)难问题,凸近似模型的求解通常会导致过收缩。本文结合加权方法和Lp范数提出了一种基于双加权L...鲁棒主成分分析(Robust principal component analysis,RPCA)模型中秩函数和L0范数的求解是非确定性多项式(Nondeterministic polynominal,NP)难问题,凸近似模型的求解通常会导致过收缩。本文结合加权方法和Lp范数提出了一种基于双加权Lp范数的RPCA模型,利用加权S p范数低秩项和加权Lp范数稀疏项分别对RPCA框架中的低秩恢复问题和稀疏恢复问题进行建模,使其更接近秩函数和L0范数最小化问题的解,提升了矩阵秩估计和稀疏估计的准确性。为了验证模型性能,本文利用图像的非局部自相似性,结合相似图像块组的低秩性与椒盐噪声的稀疏性,将双加权Lp范数鲁棒主成分分析模型应用于去除椒盐噪声过程中。定量与定性的实验结果表明,本文模型性能优于其他模型,同时奇异值过收缩分析也表明本文模型能够有效抑制秩成分的过度收缩。展开更多
By introducing an arbitrary diagonal matrix, a generalized energy function (GEF) is proposed for searching for the optimum weights of a two layer linear neural network. From the GEF, we derive a recur- sive least squa...By introducing an arbitrary diagonal matrix, a generalized energy function (GEF) is proposed for searching for the optimum weights of a two layer linear neural network. From the GEF, we derive a recur- sive least squares (RLS) algorithm to extract in parallel multiple principal components of the input covari- ance matrix without designing an asymmetrical circuit. The local stability of the GEF algorithm at the equilibrium is analytically verified. Simulation results show that the GEF algorithm for parallel multiple principal components extraction exhibits the fast convergence and has the improved robustness resis- tance to the eigenvalue spread of the input covariance matrix as compared to the well-known lateral inhi- bition model (APEX) and least mean square error reconstruction (LMSER) algorithms.展开更多
基金Supported by the National Natural Science Foundation of Chinathe Science foundation of Guangxi Educational Administration
文摘Based on the least-square minimization a computationally efficient learning algorithm for the Principal Component Analysis(PCA) is derived. The dual learning rate parameters are adaptively introduced to make the proposed algorithm providing the capability of the fast convergence and high accuracy for extracting all the principal components. It is shown that all the information needed for PCA can be completely represented by the unnormalized weight vector which is updated based only on the corresponding neuron input-output product. The convergence performance of the proposed algorithm is briefly analyzed.The relation between Oja’s rule and the least squares learning rule is also established. Finally, a simulation example is given to illustrate the effectiveness of this algorithm for PCA.
基金supported by the National Hi-Tech Research and Development Program of China("863"Project)(Grant No.2011AA040202)the National Natural Science Foundation of China(Grant No.40976114)
文摘Discriminating internal layers by radio echo sounding is important in analyzing the thickness and ice deposits in the Antarctic ice sheet.The signal processing method of synthesis aperture radar(SAR)has been widely used for improving the signal to noise ratio(SNR)and discriminating internal layers by radio echo sounding data of ice sheets.This method is not efficient when we use edge detection operators to obtain accurate information of the layers,especially the ice-bed interface.This paper presents a new image processing method via a combined robust principal component analysis-total variation(RPCA-TV)approach for discriminating internal layers of ice sheets by radio echo sounding data.The RPCA-based method is adopted to project the high-dimensional observations to low-dimensional subspace structure to accelerate the operation of the TV-based method,which is used to discriminate the internal layers.The efficiency of the presented method has been tested on simulation data and the dataset of the Institute of Electronics,Chinese Academy of Sciences,collected during CHINARE 28.The results show that the new method is more efficient than the previous method in discriminating internal layers of ice sheets by radio echo sounding data.
基金Project supported by the National Basic Research Program (973) of China (No. 2009CB320600) the National Natural Science Foun-dation of China (No. 60974057)
文摘Multiblock kernel principal component analysis (MBKPCA) has been proposed to isolate the faults and avoid the high computation cost. However, MBKPCA is not available for dynamic processes. To solve this problem, recursive MBKPCA is proposed for monitoring large scale processes. In this paper, we present a new recursive MBKPCA (RMBKPCA) algorithm, where the adaptive technique is adopted for dynamic characteristics. The proposed algorithm reduces the high computation cost, and is suitable for online model updating in the feature space. The proposed algorithm was applied to an industrial process for adaptive monitoring and found to efficiently capture the time-varying and nonlinear relationship in the process variables.
基金Supported by the Basic Research Foundation of Tsinghua National Laboratory for Information Science and Technology (TNList) the National Natural Science Foundation of China (No. 60675002)
文摘The paper analyzes the problem of blind source separation （BSS） based on the nonlinear principal component analysis （NPCA） criterion. An adaptive strong tracking filter （STF） based algorithm was developed, which is immune to system model mismatches. Simulations demonstrate that the algorithm converges quickly and has satisfactory steady-state accuracy. The Kalman filtering algorithm and the recursive leastsquares type algorithm are shown to be special cases of the STF algorithm. Since the forgetting factor is adaptively updated by adjustment of the Kalman gain, the STF scheme provides more powerful tracking capability than the Kalman filtering algorithm and recursive least-squares algorithm.
文摘近年来,鲁棒主成分分析法(Robust Principal Component Analysis,RPCA)被广泛应用到运动目标检测中,但该类方法未能有效利用运动目标的时空连续性先验,容易将动态背景误判为运动目标,且背景恢复精度不高.为此提出一种基于全变分-核回归的RPCA运动目标检测方法.该方法以RPCA为基础,利用3维全变分模型增强前景的时空连续性,去除动态背景干扰,得到清晰完整的前景.同时,利用基于扩散张量的核回归对背景的时空相关性建模,去除噪声干扰,从而精确恢复背景.在多组公开数据集上的实验结果表明,该方法在动态背景、光照变化等复杂场景中能够较为精确地检测出运动目标和恢复背景.
文摘提出一种改进的基于数据块更新的递归主元分析(recursive principal component analysis,RPCA)方法,对具有慢时变和多变量等特性的某型舰空导弹武器雷达发射机工作过程进行自适应监测。该方法在协方差矩阵的特征值分解中引入低秩奇异值分解递归方法,实现负荷矩阵和特征值矩阵的递归计算;制定了均值、方差的更新策略;给出一种基于指数加权的控制限递归算法以提高RPCA的健壮性。实验证明该方法能自适应地跟踪过程时变并实时监测故障,同时有效地降低误警率。
基金supported in part by the National Natural Science Foundation of China(Grant Nos.60172011 and 69831040)Guangxi Natural Science Foundation(Grant No.gzk0007011)the Science Foundation of Guangxi Education Bureau,China
文摘By introducing an arbitrary diagonal matrix, a generalized energy function (GEF) is proposed for searching for the optimum weights of a two layer linear neural network. From the GEF, we derive a recur- sive least squares (RLS) algorithm to extract in parallel multiple principal components of the input covari- ance matrix without designing an asymmetrical circuit. The local stability of the GEF algorithm at the equilibrium is analytically verified. Simulation results show that the GEF algorithm for parallel multiple principal components extraction exhibits the fast convergence and has the improved robustness resis- tance to the eigenvalue spread of the input covariance matrix as compared to the well-known lateral inhi- bition model (APEX) and least mean square error reconstruction (LMSER) algorithms.
