Remote Sensing Image Fusion Algorithm Based on Wavelet Coefficients

Abstract

The regional characteristics of the low frequency subband coefficients after wavelet transform and the directional region characteristics of each high frequency subband coefficient are analyzed and calculated. A new image fusion method based on the wavelet coefficient region features is proposed. For each coefficient of the low-frequency sub-band, according to the regional correlation, the fusion rule of the regional variance is adopted to determine the low-frequency fusion coefficient; for each coefficient of each high-frequency subband, according to the directional characteristic of the sub-band in which it is located, Energy fusion rules, and then determine the high-frequency fusion coefficient. Fusion experiments of multifocus images and medical images are conducted, and the fusion results are objectively evaluated with information entropy and average gradient. The experimental results show that the image fusion algorithm based on the directional characteristics of wavelet coefficients is better than the traditional fusion algorithm and has some practicality.