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 multisensor information fusion. This paper proposes a fusion rule based on the combination of PCA and ConsistencyChecking 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.

Implementation of Max Principle with PCA in image fusion for Surveillance and Navigation Application

“…The effectiveness of the proposed scheme is determined by comparing it with the author's weighted PCA method [23] and the DWT -PCA method as outlined in [25]. The images used for experimentation, the UN camp, road and boat images.…”

“…The new scheme overcomes the performance limitation of the DWT -PCA fusion algorithm [25] and the PCA weighted fusion method [23]. This is achieved by using the author's previously reported PCA based weighted fusion technique [23] along with the Select Max principle outlined in [25] to improve the fusion accuracy.…”

“…This is achieved by using the author's previously reported PCA based weighted fusion technique [23] along with the Select Max principle outlined in [25] to improve the fusion accuracy.…”

“…The low frequency images S ir and S vis are fused using the 'Select Max' principle as discussed by Zheng et al in [25]. Since the visible information is contained in the low frequency components, fusing the images by selecting the pixel values with the highest intensity gives an output image that has a very high quality as perceived by the human observer.…”

“…While this algorithm has been found to be efficient than the simple additive fusion [24] and Laplacian pyramid algorithm [8], it gives a reduced performance when compared with the hybrid DWT -PCA fusion algorithm proposed by Zheng et al [25].…”