Linear Multilayer ICA Generating Hierarchical Edge Detectors
Linear Multilayer ICA Generating Hierarchical Edge Detectors
复制标题
DOI：
10.1162/neco.2007.19.1.218
复制
发表时间：
2007
影响因子：
2.9
通讯作者：
Yoshitatsu Matsuda;K. Yamaguchi
中科院分区：
文献类型：
--
作者：
Yoshitatsu Matsuda;K. Yamaguchi
在这封信中，提出了一种新的ICA算法，线性多层ICA（LMICA）。 LMICA的每一层中都有两个阶段。其中一个是映射阶段，其中通过将更高度相关（非独立）信号与随机多维缩放网络靠近的高度相关（非独立）信号形成形成。另一个是局部 - 阶段，其中映射中的每个邻居（即高度相关）一对信号均由Maxkurt算法分开。因为在LMICA中，只有少数高度相关的对必须分开，所以它可以从自然场景中有效提取边缘检测器。我们进行了数值实验，并验证了LMICA是否从大型自然场景产生分层边缘检测器。
In this letter, a new ICA algorithm, linear multilayer ICA (LMICA), is proposed. There are two phases in each layer of LMICA. One is the mapping phase, where a two-dimensional mapping is formed by moving more highly correlated (nonindependent) signals closer with the stochastic multidimensional scaling network. Another is the local-ICA phase, where each neighbor (namely, highly correlated) pair of signals in the mapping is separated by MaxKurt algorithm. Because in LMICA only a small number of highly correlated pairs have to be separated, it can extract edge detectors efficiently from natural scenes. We conducted numerical experiments and verified that LMICA generates hierarchical edge detectors from large-size natural scenes.