Motion deblurring is a basic problem in the field of image processing and analysis. This paper proposes a new method of single image blind deblurring which can be significant to kernel estimation and non-blind deconvo...Motion deblurring is a basic problem in the field of image processing and analysis. This paper proposes a new method of single image blind deblurring which can be significant to kernel estimation and non-blind deconvolution. Experiments show that the details of the image destroy the structure of the kernel, especially when the blur kernel is large. So we extract the image structure with salient edges by the method based on RTV. In addition, the traditional method for motion blur kernel estimation based on sparse priors is conducive to gain a sparse blur kernel. But these priors do not ensure the continuity of blur kernel and sometimes induce noisy estimated results. Therefore we propose the kernel refinement method based on L0 to overcome the above shortcomings. In terms of non-blind deconvolution we adopt the L1/L2 regularization term. Compared with the traditional method, the method based on L1/L2 norm has better adaptability to image structure, and the constructed energy functional can better describe the sharp image. For this model, an effective algorithm is presented based on alternating minimization algorithm.展开更多
特征选取和子空间学习是人脸识别的关键问题。为更准确选取人脸中丰富的非线性特征,并解决小样本问题,提出了一种新的L_(2,1)范数正则化的广义核判别分析(generalized kernel discriminant analysis based on L_(2,1)-norm regularizati...特征选取和子空间学习是人脸识别的关键问题。为更准确选取人脸中丰富的非线性特征,并解决小样本问题,提出了一种新的L_(2,1)范数正则化的广义核判别分析(generalized kernel discriminant analysis based on L_(2,1)-norm regularization,L21GKDA)。利用核函数将原始样本隐式地映射到高维特征空间中,得到广义核Fisher鉴别准则,再利用一种有效变换将该非线性模型转化为线性回归模型;为了能使特征选取和子空间学习同时进行,在模型中加入了一种L_(2,1)范数惩罚项,并给出该正则化方法的求解算法。因为方法借助于L_(2,1)范数惩罚项的特征选取能力,所以它能有效地提高识别率。在ORL、AR和PIE人脸库上的实验结果表明,新算法能有效选取人脸的非线性特征,提高判别能力。展开更多
在强脉冲噪声干扰背景中,核递归最小二乘(Kernel Recursive Least Square,KRLS)算法和核递归最大相关熵(Kernel Recursive Maximum Correntropy,KRMC)算法对非线性信号预测性能严重退化,对此提出一种核递归最小平均P范数(Kernel Recursi...在强脉冲噪声干扰背景中,核递归最小二乘(Kernel Recursive Least Square,KRLS)算法和核递归最大相关熵(Kernel Recursive Maximum Correntropy,KRMC)算法对非线性信号预测性能严重退化,对此提出一种核递归最小平均P范数(Kernel Recursive Least Mean P-norm,KRLMP)算法。首先运用核方法将输入数据映射到再生核希尔伯特空间(Reproducing Kernnel Hilbert Space,RKHS)。其次基于最小P范数准则和正则化方法,推导得到自适应滤波器的最佳权向量,其降低了非高斯脉冲和样本量少的影响。然后利用矩阵求逆理论,推导得到矩阵的递归公式。最后利用核技巧得到在输入空间高效计算的滤波器输出和算法的迭代公式。α稳定分布噪声背景下Mackey-Glass时间序列预测的仿真结果表明:KRLMP算法与KRLS算法和KRMC算法相比,抗脉冲噪声能力强,鲁棒性好。展开更多
基金Partially Supported by National Natural Science Foundation of China(No.61173102)
文摘Motion deblurring is a basic problem in the field of image processing and analysis. This paper proposes a new method of single image blind deblurring which can be significant to kernel estimation and non-blind deconvolution. Experiments show that the details of the image destroy the structure of the kernel, especially when the blur kernel is large. So we extract the image structure with salient edges by the method based on RTV. In addition, the traditional method for motion blur kernel estimation based on sparse priors is conducive to gain a sparse blur kernel. But these priors do not ensure the continuity of blur kernel and sometimes induce noisy estimated results. Therefore we propose the kernel refinement method based on L0 to overcome the above shortcomings. In terms of non-blind deconvolution we adopt the L1/L2 regularization term. Compared with the traditional method, the method based on L1/L2 norm has better adaptability to image structure, and the constructed energy functional can better describe the sharp image. For this model, an effective algorithm is presented based on alternating minimization algorithm.
文摘特征选取和子空间学习是人脸识别的关键问题。为更准确选取人脸中丰富的非线性特征,并解决小样本问题,提出了一种新的L_(2,1)范数正则化的广义核判别分析(generalized kernel discriminant analysis based on L_(2,1)-norm regularization,L21GKDA)。利用核函数将原始样本隐式地映射到高维特征空间中,得到广义核Fisher鉴别准则,再利用一种有效变换将该非线性模型转化为线性回归模型;为了能使特征选取和子空间学习同时进行,在模型中加入了一种L_(2,1)范数惩罚项,并给出该正则化方法的求解算法。因为方法借助于L_(2,1)范数惩罚项的特征选取能力,所以它能有效地提高识别率。在ORL、AR和PIE人脸库上的实验结果表明,新算法能有效选取人脸的非线性特征,提高判别能力。
文摘在强脉冲噪声干扰背景中,核递归最小二乘(Kernel Recursive Least Square,KRLS)算法和核递归最大相关熵(Kernel Recursive Maximum Correntropy,KRMC)算法对非线性信号预测性能严重退化,对此提出一种核递归最小平均P范数(Kernel Recursive Least Mean P-norm,KRLMP)算法。首先运用核方法将输入数据映射到再生核希尔伯特空间(Reproducing Kernnel Hilbert Space,RKHS)。其次基于最小P范数准则和正则化方法,推导得到自适应滤波器的最佳权向量,其降低了非高斯脉冲和样本量少的影响。然后利用矩阵求逆理论,推导得到矩阵的递归公式。最后利用核技巧得到在输入空间高效计算的滤波器输出和算法的迭代公式。α稳定分布噪声背景下Mackey-Glass时间序列预测的仿真结果表明:KRLMP算法与KRLS算法和KRMC算法相比,抗脉冲噪声能力强,鲁棒性好。
