【摘要】 以夜视环境下激光助视成像和红外热成像为研究对象,提出了一种基于非向下采样Contourlet变换和分割模板的图像融合方法。通过引入脉冲耦合神经网络对两幅输入图像进行分割,利用类间方差比判断分割效果的优劣并选取分割模板。在NSCT域中,根据分割模板对激光助视成像和红外热成像的高低频系数进行分区域处理,得到低频和高频融合系数。最后对融合系数进行NSCT逆变换得到融合图像。实验结果表明本文方法能够在融合图像中最大程度呈现出夜视环境下激光助视成像的细节信息和红外热像的目标信息。更多还原

【Abstract】 Regarding laser assistant vision image and infrared thermal image under night vision environment as the research object,an image fusion algorithm based on non-subsampled contourlet transform and image segmented-template is proposed. Pulse coupled neural network is used to segment two input images,and the merits of segmentation are determined based on the between-class variance ratio,and an image segmented-template is selected. In NSCT domain,coefficients of low frequency and high frequency based on image segmented-template are processed respectively,and then coefficients of low frequency and high frequency of fused image are obtained. Finally,the fused image is obtained by performing the inverse NSCT on the fusion coefficients. The experimental results show that the proposed algorithm can best present the detail information of laser assistant vision image and target information of infrared thermal image under night vision environment.更多还原