多进制小波与自适应PCNN的可见光/毫米波图像融合

Fusion algorithm for visible and PMMW image based on multi-band wavelet and adaptive PCNN

以安检中隐匿物品检测为应用背景,提出一种基于多进制小波与自适应脉冲耦合神经网络的可见光/毫米波图像融合算法。首先可见光图与毫米波图经多进制小波分解处理,得到低频分量与高频分量。其次,低频系数采取改进区域方差融合处理,高频系数则是将子带改进拉普拉斯能量和作为PCNN中对应神经元的链接强度β,将子带八方向边缘区域能量作为PCNN的外部激励输入,经点火处理得到点火映射图,通过判决算子选取融合图像的高频系数,然后重构得到融合图。实验仿真结果分析表明,提出的融合方法在主观与客观评价上均优于现有文献中的一些典型融合方法,能获得更好的融合效果。

A new fusion algorithm which combines multi-band wavelet with adaptive pulse coupled neural network （PCNN） is pro- posed in this paper. Firstly, the original images were decomposed in the multi-band wavelet domain. Secondly, the fusion ruler for low-frequency coefficients is based on local variance, and the high-frequency coefficients is based on adaptive PCNN. The coeffi- cient improved sum of laplace energy as the PCNN link strength, and the edge-region energy of eight directions as the input of the PCNN. Finally, the fusion image is obtained by taking the inverse of multi-band wavelet transform. Lastly, the experimental results indicate that the method proposed performs better in subjective and objective assessments than a few existing typical fusion tech- niques in the literatures and obtains better fusion performance.