【摘要】 在地震属性分析技术中地震属性优化是重要的一步,主成分分析法是一种常用基于有效的线性变换的地震属性优化的方法,但对具有非线性关系地震属性数据降维效果不佳,为此提出了一种基于非线性变换的核主成分分析法。该方法通过核函数将低维输入空间映射到高维特征空间,实现了对地震属性数据的非线性到线性关系的转换,并通过主成分分析对属性优化。实验结果表明,同主成分分析法相比该方法对非线性关系的地震属性优化具有更好的效果。更多还原

【Abstract】 In the seismic attribute analysis technique,seismic attribute optimization is an important step.Principal component analysis is a commonly used method of seismic attribute optimization based on effective linear transformation,but it is not good to reduce the seismic attribute data with nonlinear relationship.A kernel principal component analysis method based on nonlinear transformation is proposed.This method maps the low-dimensional input space to the high-dimensional feature space through the kernel function,realizing the transformation from nonlinear to linear relation of seismic attribute data,and then optimizes the attribute by principal component analysis.The experimental results show that the proposed method has better effect on the seismic attribute optimization of nonlinear relation than principal component analysis.更多还原