针对混合矩阵估计算法中传统的噪声环境下基于密度的空间聚类(density-based spatial clustering of applications with noise,DBSCAN)算法需要人为设定邻域半径以及核心点数这一问题,提出双约束粒子群优化(double constrained particle...针对混合矩阵估计算法中传统的噪声环境下基于密度的空间聚类(density-based spatial clustering of applications with noise,DBSCAN)算法需要人为设定邻域半径以及核心点数这一问题,提出双约束粒子群优化(double constrained particle swarm optimization,DCPSO)算法,对DBSCAN算法的邻域半径参数进行寻优,将得到的最优参数作为DBSCAN算法的参数输入,然后计算聚类中心,完成混合矩阵估计。针对基于距离排序的源信号数目估计算法存在依靠经验参数的选取且不具备噪声点剔除能力的问题,提出了最大距离排序算法。实验结果表明,所提算法较相应的对比算法皆有提升,源信号数目估计准确率较原算法提高近40%,混合矩阵估计的误差较对比算法提升3 dB以上,且所提算法在收敛速度上优于原算法。展开更多
子空间聚类在近年来受到了大量的关注,其主要是利用谱聚类的思想学习一个表示系数矩阵以构造亲和矩阵,使用亲和矩阵获得聚类结果。众多方法采用对表示系数矩阵加以限制以保证最终得到的亲和矩阵用于聚类后得到良好的聚类效果,但这种做...子空间聚类在近年来受到了大量的关注,其主要是利用谱聚类的思想学习一个表示系数矩阵以构造亲和矩阵,使用亲和矩阵获得聚类结果。众多方法采用对表示系数矩阵加以限制以保证最终得到的亲和矩阵用于聚类后得到良好的聚类效果,但这种做法会降低亲和矩阵的表示能力。本文提出低秩稀疏亲和矩阵子空间聚类算法,直接对亲和矩阵进行约束以提高表示系数矩阵的表示能力。文章给出了算法的优化过程,验证了结果的块对角性质,在不同数据集上的实验证明了方法的有效性。Subspace clustering has received a lot of attention in recent years, which mainly uses the idea of spectral clustering to learn a representation coefficient matrix to construct an affinity matrix, and uses the affinity matrix to obtain clustering results. Many methods use the restriction of the representation coefficient matrix to ensure that the final affinity matrix is used for clustering to obtain good clustering results, but this practice will reduce the representation ability of the affinity matrix. In this paper, a low-rank sparse affinity matrix subspace clustering algorithm is proposed to directly constrain the affinity matrix to improve the representation ability of the representation coefficient matrix. The optimization process of the algorithm is presented, and the block diagonal property of the results is verified. Experiments on different data sets prove the effectiveness of the method.展开更多
子空间聚类是聚类来源于底层子空间的数据的一个高效的方法。在近些年,基于谱聚类的方法成为了最受欢迎的子空间聚类方法之一。新近提出的自适应图卷积子空间聚类方法受图卷积网络的启发,使用图卷积技术去设计了新的特征提取的方法和系...子空间聚类是聚类来源于底层子空间的数据的一个高效的方法。在近些年,基于谱聚类的方法成为了最受欢迎的子空间聚类方法之一。新近提出的自适应图卷积子空间聚类方法受图卷积网络的启发,使用图卷积技术去设计了新的特征提取的方法和系数矩阵的约束,取得了优异的效果。但其需要重构系数矩阵满足对称和非负的条件,这会限制重构系数矩阵的表示能力。为了克服这一缺陷,本文改为直接约束由重构系数矩阵生成的亲和矩阵,亲和矩阵天然具有对称和非负的性质,进而设计了亲和矩阵图卷积子空间聚类算法。不仅克服了求解模型的困难之处,还进行了对比实验在四个基准数据集上以此论证本文方法的有效性。Subspace clustering is an efficient method for clustering data derived from the bottom level subspace. In recent years, spectral clustering based methods have become one of the most popular subspace clustering methods. The recently proposed adaptive graph convolution subspace clustering method is inspired by graph convolutional networks and uses graph convolution techniques to design new feature extraction methods and constraints on coefficient matrices, achieving excellent results. But it requires the reconstruction coefficient matrix to satisfy symmetric and non negative conditions, which limits the representational power of the reconstructed coefficient matrix. To overcome this limitation, this paper proposes to directly constrain the affinity matrix generated from the reconstructed coefficient matrix, which naturally has symmetric and non negative properties. Therefore, an affinity matrix graph convolution subspace clustering algorithm is designed. Not only did it overcome the difficulties in solving the model, but it also conducted comparative experiments on four benchmark datasets to demonstrate the effectiveness of the proposed method.展开更多
文摘子空间聚类在近年来受到了大量的关注,其主要是利用谱聚类的思想学习一个表示系数矩阵以构造亲和矩阵,使用亲和矩阵获得聚类结果。众多方法采用对表示系数矩阵加以限制以保证最终得到的亲和矩阵用于聚类后得到良好的聚类效果,但这种做法会降低亲和矩阵的表示能力。本文提出低秩稀疏亲和矩阵子空间聚类算法,直接对亲和矩阵进行约束以提高表示系数矩阵的表示能力。文章给出了算法的优化过程,验证了结果的块对角性质,在不同数据集上的实验证明了方法的有效性。Subspace clustering has received a lot of attention in recent years, which mainly uses the idea of spectral clustering to learn a representation coefficient matrix to construct an affinity matrix, and uses the affinity matrix to obtain clustering results. Many methods use the restriction of the representation coefficient matrix to ensure that the final affinity matrix is used for clustering to obtain good clustering results, but this practice will reduce the representation ability of the affinity matrix. In this paper, a low-rank sparse affinity matrix subspace clustering algorithm is proposed to directly constrain the affinity matrix to improve the representation ability of the representation coefficient matrix. The optimization process of the algorithm is presented, and the block diagonal property of the results is verified. Experiments on different data sets prove the effectiveness of the method.
文摘子空间聚类是聚类来源于底层子空间的数据的一个高效的方法。在近些年,基于谱聚类的方法成为了最受欢迎的子空间聚类方法之一。新近提出的自适应图卷积子空间聚类方法受图卷积网络的启发,使用图卷积技术去设计了新的特征提取的方法和系数矩阵的约束,取得了优异的效果。但其需要重构系数矩阵满足对称和非负的条件,这会限制重构系数矩阵的表示能力。为了克服这一缺陷,本文改为直接约束由重构系数矩阵生成的亲和矩阵,亲和矩阵天然具有对称和非负的性质,进而设计了亲和矩阵图卷积子空间聚类算法。不仅克服了求解模型的困难之处,还进行了对比实验在四个基准数据集上以此论证本文方法的有效性。Subspace clustering is an efficient method for clustering data derived from the bottom level subspace. In recent years, spectral clustering based methods have become one of the most popular subspace clustering methods. The recently proposed adaptive graph convolution subspace clustering method is inspired by graph convolutional networks and uses graph convolution techniques to design new feature extraction methods and constraints on coefficient matrices, achieving excellent results. But it requires the reconstruction coefficient matrix to satisfy symmetric and non negative conditions, which limits the representational power of the reconstructed coefficient matrix. To overcome this limitation, this paper proposes to directly constrain the affinity matrix generated from the reconstructed coefficient matrix, which naturally has symmetric and non negative properties. Therefore, an affinity matrix graph convolution subspace clustering algorithm is designed. Not only did it overcome the difficulties in solving the model, but it also conducted comparative experiments on four benchmark datasets to demonstrate the effectiveness of the proposed method.
