基于双特征量和NSCT的多波段SAR图像融合算法

Multi-bands SAR image fusion algorithm based on dual-features and NSCT

  • 摘要: 不同波段的SAR图像进行融合,能够综合不同波段SAR图像的信息,得到包含更多信息的SAR图像。针对传统融合方法不能很好地处理SAR图像中边缘及纹理等细节信息的问题,该文提出一种基于双特征量的在非下采样Contourlet变换域融合的方法。对NSCT分解后的高频分量提出了一种基于区域能量和梯度两个特征量,采用平均与选择相结合的规则来计算高频子带系数。实验结果表明,该方法得到的融合图像在客观评价上要优于传统融合方法,这说明该方法在综合了不同波段SAR图像信息的同时能更好地保持细节信息。

     

    Abstract: Fusion of SAR images in different bands can synthesize the information of SAR images in different bands and obtain SAR images with more information. Aiming at the problem that the traditional fusion method cannot deal with the details such as edge and texture in SAR image well, a fusion method based on dual-features in non-subsampling contourlet transform (NSCT) domain is proposed in this paper. For the high-frequency components, a new method based on the region energy and the region gradient is proposed, and the high-frequency coefficients are obtained by the combination of the average and selection rules. Finally, the fusion image is obtained by NSCT inverse transformation. The experimental results show that the fusion image obtained by this method is better than the traditional fusion methods in objective evaluation, which shows that proposed method can synthesize the information of SAR images in different bands and keep details well.

     

/

返回文章
返回