一种基于区域特性的红外与可见光图像融合算法

提出了一种基于区域分割和à trous小波变换的红外与可见光图像融合算法.首先,对红外与可见光图像进行区域分割及区域关联,并按关联映射图所划分区域提取红外与可见光图像的的能量信息及梯度信息；然后,对红外与可见光图像进行多尺度à trous小波变换分解,分解后的低频部分按照文中所提出的区域能量比和区域清晰比指标进行区域融合,高频部分采用绝对值取大算子进行融合；最后进行重构得到融合图像.结果表明,该算法既可保持可见光图像的光谱信息,又可有效获取红外图像的热目标信息.

Fusion Algorithm of Infrared and Visible Images Based on Region Feature

An fusion algorithm of infrared and visible images based on region segmentation and à trous wavelet transform was proposed.Region segmentation and region joint were done for the infrared and visible images.Energy information and gradient information of the infrared and visible images were extracted according to the regions generated with joint map.And multiscale decomposition of the infrared and visible images was done with the à trous wavelet transform.The low-frequency coefficients were fused with the region-based rules according to the ratio of region energy(RRE) and ratio of region sharpness (RRS) presented,and the high-frequency coefficients were fused with larger absolute value operator.The fused coefficients were reconstructed to obtain the fusion image.Experimental results show that the presented algorithm can hold spectrum information of the visual image,and effectively get hot object information of the infrared image.