A SAR-optical Image Fusion Method Based on DT-CWT

Abstract

Fusion method based on Discrete Wavelet Transform (DWT) has been wide used in Multisource image fusion. However, some shortcomings such as the shift sensitivity and the poor-direction-selectivity of DWT afiected the quality of the fused image. In this paper, a fusion method based on Dual-tree Complex Wavelet (DT-CWT) is proposed, which can be used for integrated map from optical image and SAR image. DT-CWT is an extension of DWT, it provides a simple mathematical framework, multiresolution and multi-direction. Fusion method based on DT-CWT is more efiective than DWT because the direction selectivity and translational stability is better than DWT. This paper proposed a high frequency component fusion method based on the outline matching rule. Compared with the traditional weighting fusion method, this method will get a better fusion quality. At the same time, the calculation amount is less than the fusion method based on artiflcial intelligence. A simulation experiment has been accomplished in order to verify the efiects of this method. The simulation result shows that there is a good result in SAR-optical fusion in urban area.