Image fusion algorithm based on nonsubsampled contourlet transform and regional fractal dimension

As an important problem in the course of image fusion, the extraction of useful information from source images affects the fusion result directly, and it is all along one of the research focuses and difficult ponits of image fusion. A novel image fusion method for infrared and visible images based on nonsubsampled contourlet transform and fractal dimension is proposed. The regional properties, size of regions, intensity of edges and the strength of texture are all described by the regional fractal dimensions in different scales. Through selecting coefficients from two source images based on regional fractal dimension, the low frequency subband coefficients of the source images are fused. The bandpass directional coefficients are fused using different strategies based on the matching operator proposed. Compared with other methods, experimental results show that the proposed algorithm is effective and feasible.