摘要
为了获得多聚焦图像的最佳融合效果,得到更多细节信息,分析了相关的图像融合算法,选择变换域内融合方法对图像进行融合。对两张待融合图像使用小波变换进行分层(小波基选择sym4,分解层数为3层),低频系数采用加权平均法融合。高频系数使于Canny算子的边缘检测,对检测到的边缘图像采用最大系数法进行融合。对加权平均法、小波变换(低频平均、高频最大系数融合)、及改进后算法进行仿真分析,直观视觉及参数分析中均表明改进后算法图像融合效果更好,细节保留、轮廓清晰度显著提高。
In order to obtain the best fusion effect of multi-focus image and get more details,the relevant image fusion algorithms are analyzed,and the fusion method in transform domain is selected for image fusion.Two images to be fused are stratified by wavelet transform(sym4is chosen as the wavelet basis and 3levels are decomposed),and the low frequency coefficients are fused by weighted average method.The high frequency coefficient makes the edge detection of Canny operator.The maximum coefficient method is used to fuse the detected edge images.The weighted averaging method,wavelet transform(low-frequency averaging,high-frequency maximum coefficient fusion)and the improved algorithm are simulated and analyzed.Visual vision and parameter analysis show that the improved algorithm has better image fusion effect,better details retention and better contour clarity.
作者
王园园
张娜
韩美林
Wang Yuanyuan;Zhang Na;Han Meilin(College of Electronic Information and Electrical Engineering,Shangluo University,Shangluo 726000,China)
出处
《电子测量技术》
2019年第17期148-151,共4页
Electronic Measurement Technology
基金
陕西省商洛市科技局项目(sk2017-42)资助
