Image fusion technique based on non-subsampled contourlet transform and adaptive unit-fast-linking pulse-coupled neural network

摘要

A new image fusion technique based on non-subsampled contourlet transform (NSCT) and adaptive unit-fast-linking pulse-coupled neural network (PCNN) is presented. By using NSCT, multi-scale and multi-direction sparse decompositions of the source images are performed. Then, the basic PCNN model is improved to be an adaptive unit-fast-linking PCNN model, which synthesises the advantages of both unit-linking PCNN and fast-linking PCNN. The novel PCNN model utilises the clarity of each pixel in images as the linking strength b; moreover, the time matrix T of the sub-images can be obtained via the synchronous pulse burst property. Finally, the sub-images are fused by analysing the time matrix T and linking strength b. The experimental results show that the proposed approach is better than some current methods.