基于DT-CWT和自适应PCNN模型的多聚焦图像融合方法研究  

Research on Multi-focus Images Fusion Method Based on DT-CWT and Adaptive PCNN Model

作　　者：张腾敏 林斯乐 朱胜[1] 洪培瑶 ZHANG Teng-min;LIN Si-le;ZHU Sheng;HONG Pei-yao(Jinshan College, Fujian Agriculture and Forestry University, Fuzhou 350002, China)

机构地区：[1]福建农林大学金山学院,福州350002

出　　处：《唐山学院学报》2021年第6期16-22,69,共8页Journal of Tangshan University

基　　金：福建省教育厅中青年教师教育科研项目(JAT201000);福建农林大学金山学院科研项目(kx210318)。

摘　　要：针对传统图像融合方法在多聚焦图像融合中存在细节丢失、边缘模糊和焦点不清楚等问题,提出一种基于双树-复小波变换(DT-CWT)优化显著性测度和自适应脉冲耦合神经网络(PCNN)模型的多聚焦图像融合方法。首先,将两张聚焦区域不同的输入图像使用双树-复小波分解成低频子带和高频子带;然后,对低频子带采用基于显著性测度的度量方法计算小波融合系数,对于高频子带,采用自适应PCNN模型计算触发时间来选取高频融合子带;最后,通过双树-复小波逆变换重构得到融合结果。与其他融合方法进行对比,结果表明,基于文章所提方法的融合图像更加自然清晰,具有较高的边缘保持度,同时保留了更多的细节信息,因此,此方法可以大大提高图像质量。Aiming at the problems in the traditional image fusion method,such as loss of details,blurred edges and unclear focus in multi-focus image fusion,a new method is proposed based on the dual-tree complex wavelet transform(DT-CWT)optimizing significance measure and adaptive pulse coupled neural network(PCNN)model.First of all,two input images with different focus areas are decomposed into low-frequency sub-bands and high-frequency sub-bands with dual-tree complex wavelet.Then,for the low-frequency sub-bands,the significance measure is used to calculate the wavelet fusion coefficient;And for the high frequency sub-band,the adaptive PCNN model is used to calculate the trigger time to select the high-frequency fusion sub-band.Finally,the fusion result is obtained through the reconstruction of dual-tree complex wavelet transform.The experiment results show that the fusion image with the above method is more natural and clear,with higher edge retention,and more detailed information retained.Therefore,the method proposed in this article can greatly improve the image quality.

关 键 词：多聚焦图像 融合方法 双树-复小波变换 显著性测度 自适应PCNN模型 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]