基于残差密集网络的红外与可见光图像配准

Infrared and visible image registration based on residual dense network

针对难以提取红外与可见光图像中有效特征的问题,提出一种基于残差密集网络的红外与可见光图像配准方法。首先,利用基于残差密集网络设计的特征提取网络来提取图像对的分层特征。其次,利用掩码预测网络产生的掩码来对特征进行加权映射,从而突出图像中的重要特征。最后,这些重要特征通过单应性网络得到变换矩阵,从而实现图像配准。实验结果表明,残差密集网络的引入可提升配准效果,且评估指标ACE提升了21.5%。

Aiming at the problem that it is difficult to extract effective features from infrared and visible images,an infrared and visible image registration method based on the residual dense network is proposed.Firstly,the hierarchical features of image pairs are extracted by using the feature extraction network designed based on the residual dense network.Secondly,the mask generated by the mask prediction network is used to weighted map the features,which highlights the important features in the image.Finally,using these important features,the transformation matrix is obtaineb through a homography network,so as to realize image registration.The experimental results show that the introduction of the residual dense network can improve the registration effect,and the evaluation index ACE is improved by 21.5%.