论文部分内容阅读
在奇异值字典学习方法的基础上,结合主成分分析方法提出了主成分分析字典学习方法.该方法取代了奇异值分解(KSVD)方法中对误差项直接进行SVD分解来更新原子,取而代之的是通过对误差项进行PCA分解,提取其主成分作为字典中原子的更新.仿真结果表明,与KSVD字典学习方法相比,所提出的方法字典学习效果更好,对训练样本的表达误差更小,学习字典更能表达训练样本的特征.
Based on the learning method of singular value dictionary, this paper proposes a method of principal component analysis (DPCA) based on Principal Component Analysis (DPCA), which replaces the singular value decomposition (KSVD) method to directly update the atom by using SVD to replace the error term The PCA decomposition of the error term is used to extract the principal component of the dictionary as the update of the atom in the dictionary.The simulation results show that compared with the KSVD dictionary learning method, the proposed method has better learning performance and less error on the training samples, Learning dictionaries can better express the characteristics of training samples.