摘要
针对图像融合算法存在细节不清晰缺点,本文提出了一种改进的基于局部特征的图像融合算法。首先,对源图进行小波变换,得到低频与高频系数。然后,对于低频系数,通过对局部特征矩阵进行奇异值分解得到特征值,由特征值确定加权系数;对于高频系数,由局部匹配度和特征来确定高频融合系数,有效保留源图边缘信息。最后,通过小波逆变换得到融合图像。实验结果表明,相比于其他融合算法,本文算法融合图像客观指标更好,细节更清晰。
In view of the shortcoming of image fusion based on discrete wavelet transform(DWT)with unclear textural information,an improved image fusion based on local features is proposed in this paper.First,the visible light and infrared images are decomposed into low frequency coefficients and high frequency coefficients.Second,For the low frequency coefficients,eigenvalues are obtained by singular value decomposition of local characteristic matrix,then the weighted coefficients are determined by eigenvalues.For the high frequency coefficients,the fused high frequency coefficients are determined by local matching degree and feature,and the edge information of source images is reserved effectively.Finally,the fused image is obtained by inverse DWT.Experimental results demonstrate that the objective index of the proposed method is better and fused image is more detailed,compared with other fusion algorithms.
作者
王立娟
WANG Li-juan(Nanjing University of Science and Technology of Zijin College,College of electronic engineering and Optoelectronic Technology,Nanjing,Jiangsu 210000)
出处
《新型工业化》
2020年第3期96-99,共4页
The Journal of New Industrialization
关键词
图像融合
小波变换
局部特征
局部匹配度
Image fusion
Wavelet transform
Local features
Local matching degree
