基于快速清晰度估计的多聚焦图像融合

为了准确地估计源图像的清晰区域,提高多聚焦图像融合的效率,提出了一种新的基于清晰度估计的图像融合方法。首先利用基于离散小波的清晰度估计方法获取源图像的聚焦区域;然后使用均值滤波和空洞填充进一步优化该聚焦区域;最后结合清晰度估计和相似性特性,将不同聚焦区域合并生成融合图像。该方法获得的融合图像在客观评价和主观质量上都优于以往基于清晰度的图像融合方法。...

Full description

Saved in:
Bibliographic Details
Published in计算机应用研究 Vol. 33; no. 3; pp. 896 - 898
Main Author 费春 张萍 李建平
Format Journal Article
LanguageChinese
Published 电子科技大学 计算机科学与工程学院,成都,610054%电子科技大学 光电信息学院,成都,610054 2016
Subjects
图像融合
多聚焦
清晰度估计
相似性
sharpness estimation
清晰度估计
similarity characteristics
multi-focus
image fusion
图像融合
多聚焦
相似性
Online AccessGet full text

Cover

More Information
Summary:为了准确地估计源图像的清晰区域,提高多聚焦图像融合的效率,提出了一种新的基于清晰度估计的图像融合方法。首先利用基于离散小波的清晰度估计方法获取源图像的聚焦区域;然后使用均值滤波和空洞填充进一步优化该聚焦区域;最后结合清晰度估计和相似性特性,将不同聚焦区域合并生成融合图像。该方法获得的融合图像在客观评价和主观质量上都优于以往基于清晰度的图像融合方法。
Bibliography:51-1196/TP
multi-focus; sharpness estimation; image fusion; similarity characteristics
Fei Chun,Zhang Ping,Li Jianping（a. School of Computer Science ＆ Engineering, b. School of Optoelectronic Information, University of Electronic Science ＆ Technology of China, Chengdu 610054, China）
This paper proposed a novel method multi-focus image fusion based on sharpness estimation in order to accurately estimate the sharpness area of source images and improved the efficiency of multi-focus image fusion. First,the proposed method obtained the focus areas from source images by using a sharpness estimation method based on discrete wavelet transform.Then,it used mean filter and hole-filling to optimize the focus areas. Finally,the proposed method selected different focus areas to generate fusion image combining sharpness estimation and similarity characteristics. The experimental results show that the proposed method outperforms to conventional fusion methods based on sharpness estimation in terms of objective evaluation and
ISSN:1001-3695
DOI:10.3969/j.issn.1001-3695.2016.03.058