ID 原文 译文
8144 考虑到不同特征对总体云关联度的贡献不同,根据特征类内和类间散度对云关联度矩阵进行加权。 Considering the different characteristics of contribution to the overall cloud correlation degree, according to the characteristics of the divergence between class and class of cloud weighted correlation matrix.
8145 最后,通过求取待识别特征向量与各故障模式的云关联序列,实现对电路故障的识别诊断。 Finally, by calculating which cloud the feature vector and each failure mode correlation sequence, to realize the recognition of circuit fault diagnosis.
8146 实例分析表明,所提方法在容差条件下对电路软故障有较高的诊断效率。 The example analysis show that the proposed method under the condition of the tolerance of circuits soft fault has high diagnostic efficiency.
8147 高阶全变分图像去噪方法利用图像方向导数的可分L1范数,构建优化方程进行图像去噪,可以在去除图像噪声的同时有效保留图像中的细节信息。 High order variational image denoising method using the directional derivative of the image can be divided into the L1 norm, build optimized equation in image denoising, can remove the noise of the image at the same time keep the details of images effectively.
8148 然而传统高阶全变分方法计算复杂度较高、耗时较长。 However traditional higher-order total variational method high computation complexity, time consuming is longer than the other.
8149 针对此问题,提出了一种基于增广拉格朗日乘子的快速高阶全变分图像去噪方法。 Aiming at this problem, this paper proposes a fast based on augmented Lagrange multiplier high order variational image denoising methods.
8150 首先,利用Huber方程重建高阶全变分优化方程; First of all, the reconstruction of Huber equation all higher order variational optimization equation;
8151 其次,通过添加辅助变量及引入拉格朗日乘子,将优化方程转换为两个较易求解的子问题进行交替最小化迭代求解。 Second, by adding auxiliary variables and the introduction of Lagrange multiplier, the optimization equation solving more easily converted into two subproblems to alternate iteration solving minimization.
8152 实验证明,在相同条件下,与传统方法相比,基于增广拉格朗日乘子的高阶全变分图像去噪方法可以大幅提高运算速度,并且能在去除图像噪声的同时更好地保留图像边缘、纹理、细节等信息,获得视觉效果更好的去噪图像。 Experiments show that under the same conditions, compared with the traditional method, based on augmented Lagrange multiplier of higher order variational image denoising method can greatly improve the computing speed, and can remove the image noise and better retention of image edge, texture and details information, obtain better visual effect of image denoising.
8153 为了在低运算量下提高均匀圆阵中二维测角的精度,提出一种基于圆阵模式空间和多重旋转不变子空间算法的二维波达方向估计算法。 In order to improve the uniform circular array under the low computational complexity in the two-dimensional Angle measuring accuracy, a model-based circle pattern space and multiple rotation invariant subspace algorithm of 2-d doa estimation algorithm.