Effective Image Fusion Rules Of Multi-scale Image Decomposition

Abstract

Image fusion rule is the key that influences the quality of image fusion which is a primary branch of multi-sensor information fusion. This paper proposes a fusion rule based on the combination of PCA and consistency checking and applies it to multi-scale image fusion algorithm. The method applies the PCA method to the approximation image decomposed by multi-scale and adopts "select max" on detail image, the consistency of which is later checked with fusion decision. The proposed rule overcomes the instability and incontinuity of information occurring in the image fusion decomposed by multi-scale before. The experimental results show that better than the traditional rules, this method can effectively improve the quality of multi-sensor image fusion, and is especially applicable to images fusion of different sensors.