Multi-directional Laplacian Pyramid Image Fusion Algorithm

Abstract

The traditional Laplacian pyramid fusion method cannot effectively reflect the edge information of the fused image and the poor contrast. A Laplacian pyramid algorithm with multi-directional joint averaging is proposed. First, the input image is decomposed using the Laplacian pyramid. Secondly, for the high frequency coefficient, the rule of extracting the multi-directional feature map of the joint average fusion is adopted; the low frequency coefficient adopts the regional average gradient adaptive weighted fusion rule. Finally, the fused image is obtained by the reconstruction process of the inverse Laplacian pyramid transform. Experimental results show that the algorithm can effectively reflect the edge information of the fused image and rich background details.