本文针对传统离散小波变换（ DWT）在图像融合中细节丢失的问题，提出了一种基于双树复小波变换（DT-CWT）的低频域区域能量取大和高频域自适应脉冲耦合神经网络（PCNN）图像融合算法（简称 DC-SA-PCNN）。实验结果显示，采用同样的融合规则， DT-CWT融合图像的互信息量 MI、边缘保持度 QAB/F、融合积 MQ＝MI×QAB/F均高于 DWT融合图像，基于自适应 PCNN算法获得的融合图像具有更优的 MI、QAB/F、MQ指标。结果表明，DC-SA-PCNN算法有效地综合了红外图像和可见光图像中的信息，融合图像更加全面地携带了源图像中的有效信息特征。

Aiming at the problem of loss of detail in image fusion by traditional discrete wavelet transform (DWT), this paper proposes an adaptive pulse-coupled neural network(PCNN) for large and high frequency domains in a low frequency domain based on the double-tree complex wavelet transform(DT-CWT) image fusion algorithm(referred to as DC-SA-PCNN). The experimental results show that the mutual information MI, edge retention QAB/F, and fusion product MQ＝MI×QAB/F of the DT-CWT fusion image are higher than that of the DWT fusion image, and that fusion based on the adaptive PCNN algorithm using the same fusion rule image has better MI, QAB/F, and MQ indicators. These results indicate that the DC-SA-PCNN algorithm effectively combines the information in the infrared and visible image, and the fused image carries more sufficient effective information features in the source image.