Image fusion via wavelet transform based on local contrast

In this article, images obtained by multi-sensor use multi-resolution analysis of image fusion methods based on wavelet transform. Two source images obtained high frequency and low frequency components after wavelet transform. Using fusion rules with criteria based on local variance 171 to obtain low-frequency component. using an adaptive algorithm with local contrast of images to obtain the high-frequency component; at last, using inverse wavelet transform to rebuild a fusion image with useful information from source image. Results show that compared with traditional algorithm and general algorithm, fusion image obtained by the proposed algorithm increases Peak-to-peak Signal-to-Noise Ratio by 12.11% and 8%, reduces root mean square error by 69.65% and 59.80%, increases correlation coefficient by 0.95% and 0.52%.