A New Multi-focus Image Fusion Algorithm Based on BEMD and Improved Local Energy

摘要

A multi-focus image fusion method, which is based on bidimensional empirical mode decomposition and improved local energy algorithm, is presented in this paper. First, source image is decomposed by bidimensional empirical mode decomposition. Then maximum criterion combined with weighted average fusion rule based on local energy is applied to bidimensional intrinsic mode function components of the corresponding frequency segment. If phase of bidimensional intrinsic mode function coefficients decomposed by bidimensional empirical mode decomposition on two source images is same, local energy maxima criterion is used in frequency coefficients of fused image, elseif corresponding phase is opposite, bidimensional intrinsic mode function coefficients of fused image is determined by weighted average method based on local energy. Finally, fusion result is received by inverse bidimensional empirical mode decomposition transform on fusion coefficient. Simulation shows that the proposed algorithm is significantly outperforms the traditional methods, such as maximum criterion, weighted average and wavelet fusion rules.