With the development of fusion technology and the growth of the wavelet theory,wavelet transform uses its excellent time and frequency localization,excels in the area of fusion.On the basis of the wavelet theory,we propose a new algorithm which combines wavelet transform with the pulse coupled neural networks.Firstly,we perform a wavelet multiscale decomposition of two original images which have been registered.Secondly, we propose a novel fusion rule based on the improved PCNN.The key point in this paper is that in allusion to the feature of the high frequency domain and the low frequency one,using various methods to each frequency domain,respectively.Finally,it can obtain a fused image by taking inverse wavelet transform to reconstruct images.The results of simulation and quantifying evaluation show that this algorirhm can preserve more useful information from the original images effectively,and enhance the quality of the fused image.