Image fusion algorithm based on adaptive pulse coupled neural networks in curvelet domain

Using the fast discrete curvelet transform, an image fusion algorithm based on adaptive pulse coupled neural networks (PCNNs) is proposed. PCNN is built in each high-frequency subband to simulate the biological activity of human visual system. Support vector machine is employed to achieve support values which represent subband features and then will be imported to motivate the neurons. The first firing time of each neuron is presented as the salience measure. Compared with traditional algorithms where the linking strength of each neuron is set as constant or always changed according to features of each pixel, in our algorithm, the linking strength as well as the linking range is determined by the prominence of corresponding low-frequency coefficients, which not only reduces the calculation of parameters but also flexibly makes good use of global features of images. Experimental results indicate superiority of the proposed algorithm in terms of visual effect and objective evaluations.