A Novel Surface Interpolation Approach for Bidimensional Empirical Mode Decomposition

The surface interpolation of the scattered points is the key technology in Bidimensional Empirical Mode Decomposition (BEMD) algorithm that determines the quality of Bidimensional Intrinsic Mode Functions (BIMF). The traditional interpolation method commonly has high complexity and often results in boundary effects. A new interpolation method based on the nonparametric kernel regression (KR) with the Nadaray-Watson filter is proposed. The accuracy of the interpolation can be modified by the filter parameter, and the direction of the filter is adaptive to maintain the characteristic of the result surface. Simulation results show that the surface interpolation based on KR can reduce the complexity of the algorithm and avoid the boundary effects, improving the performance of BEMD in edge detection.