光学学报, 2017, 37 (10): 1010002, 网络出版: 2018-09-07   

基于剪切波变换和邻域结构特征的红外与可见光图像融合 下载: 902次

Fusion of Infrared and Visible Images Based on Shearlet Transform and Neighborhood Structure Features
作者单位
空军工程大学航空航天工程学院, 陕西 西安 710038
摘要
针对传统融合方法融合后目标轮廓模糊和细节不突出等问题,提出一种在剪切波框架下基于邻域结构特征的红外与可见光图像融合算法。通过剪切波变换对源图像进行分解得到与源图像同尺寸的高频和低频子带系数;为防止融合后图像边缘模糊,对低频子带系数采用几何距离与能量距离加权的融合规则,对高频子带系数采用灰度差异与梯度距离加权的融合规则来更好地保留源图像的细节信息;经剪切波逆变换得到融合后图像。结果表明,本文算法能有效地提取红外目标信息和保持可见光图像信息;在保留图像轮廓信息的基础上,凸显目标信息,有效地改善图像融合效果。
Abstract
In view of the problems that the target contour of the fused image is fuzzy and its details are not highlighted by the use of traditional fusion algorithms, an infrared and visible image fusion algorithm based on the shearlet frame and neighborhood structure features is proposed. The shearlet transform is used to decompose the source images to get the subbands coefficients of high frequency and low frequency with the same size as the original images. Then, in order to prevent the edge of the fusion image from blurring after fusion, a fusion rule based on geometrical distance combined with energy distance is adopted in low frequency subband coefficients. Moreover, a fusion strategy based on gray difference and gradient distance weighting is used to fuse high frequency subband coefficients for keeping the details of the images better. Finally, the fusion image is obtained by shearlet inverse transformation. Results show that the proposed algorithm can effectively extract the target infrared information and keep the visible image information. On the basis of retaining the image profile information,the proposed algorithm can highlight the target information, and improve the image fusion effect effectively.

丁文杉, 毕笃彦, 何林远, 凡遵林, 吴冬鹏. 基于剪切波变换和邻域结构特征的红外与可见光图像融合[J]. 光学学报, 2017, 37(10): 1010002. Wenshan Ding, Duyan Bi, Linyuan He, Zunlin Fan, Dongpeng Wu. Fusion of Infrared and Visible Images Based on Shearlet Transform and Neighborhood Structure Features[J]. Acta Optica Sinica, 2017, 37(10): 1010002.

本文已被 6 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!