Fusion of Multi-sensor Images Based on PCA and Self-Adaptive Regional Variance Estimation

Abstract

An algorithm is presented for exploiting the properties of the lifting wavelet transform for multi-sensor image fusion. The method includes adaptive fusion arithmetic based on principal component analysis (PCA) and self-adaptive regional variance estimation. Characteristics of the wavelet coefficients are used to adaptively select fusion rules. A weighting method based on PCA is applied to low-frequency image components, and the regional variance estimation is applied to high-frequency components including edges and details of the original image. Experiments reveal that the methods are effective for multi-focus, visible-light, and infrared image fusion. Compared with traditional algorithms, the new algorithm not only improves the amount of preserved information and clarity, but also increases the correlation coefficient between the fused and source images.