Multi-focus Image Fusion Based on Wavelet Coefficients Selection and Reconstruction

For the fusion problem of the multifocus image of the same scene, a novel algorithm is proposed on the basis of the selection and reconstruction of the coefficient in wavelet domain. For choosing the low frequency subband coefficients, a scheme based on the regional standard deviation and mean of the high frequency sub-images, which can combine with low frequence and high frequence subband coefficients, is provided. To reconstruct the high-frequency sub-band coefficient, a novel method based on the structure tensor analysis is employed. In the method, the high frequency subband coefficients in horizontal and vertical orientation are reconstructed by using eigenvalue analysis of the weighted structure tensor. The method can obtain more detailed information and enhance contrast. Experimental results show that, compared to traditional algorithm based on wavelet transform, the proposed algorithm can get a better fusion results in terms of both visual effects and quantitative indicators.