基于MSSTO与NSCT变换的可见光与红外图像增强融合

发布时间：2018-05-20 23:29

本文选题：图像融合 + 非下采样轮廓波变换　；参考：《控制与决策》2017年02期

【摘要】：针对红外与可见光图像融合结果中边缘区域失真严重、对比度差的问题,提出一种基于多尺度顺序翻转算子(MSSTO)和非下采样轮廓波变换(NSCT)的图像增强融合算法.首先,采用NSCT将图像分解成高低频系数;其次,利用MSSTO从低频系数中提取出有效的亮、暗信息,并将其注入到融合低频系数中以合成最终低频系数;再次,高频系数采用局部空间频率加权(LFSW)与区域能量取大的融合方案;最后,对合成的高低频系数进行反NSCT得到融合图像.实验结果验证了所提出算法的有效性.
[Abstract]:Aiming at the serious edge distortion and poor contrast in infrared and visible image fusion, an image enhancement fusion algorithm based on multi-scale sequential flipping operator (MSSTO) and non-downsampling profilometry transform (NSCT) is proposed. Firstly, the image is decomposed into high and low frequency coefficients by NSCT. Secondly, the effective bright and dark information is extracted from the low frequency coefficients by MSSTO and injected into the fusion low frequency coefficients to synthesize the final low frequency coefficients. The high frequency coefficients are fused with the local spatial frequency weighted NSCT and the region energy is increased. Finally, the fusion image is obtained by inverse NSCT of the synthesized high and low frequency coefficients. Experimental results show that the proposed algorithm is effective.
【作者单位】：西北工业大学自动化学院;
【基金】：国家自然科学基金重点项目(61135001) 西安市科技计划项目(CXY1436(9);CXY1350(2))
【分类号】：TP391.41

【相似文献】