A multisensor image fusion algorithm based on PCNN

Based on the principle of pulse-coupled neural network (PCNN), a novel algorithm for multisensor image fusion is presented. Firstly a contrast pyramid decomposition of source images is performed, and then the contrast pyramids are used as the input of PCNN. The contrast is selected based on the number of output pulse of PCNN to realize image fusion. The novel algorithm utilizes the global feature of source images because PCNN has the global coupled and pulse synchronization characteristics. It accords with the physiological characteristic of human visual neural system. The novel algorithm is applied to fuse charge-coupled device (CCD) and synthetic aperture radar (SAR) images, and the fusion result is compared with those of some other fusion methods through some performance evaluation measures for fusion effect. Comparison results show that the novel fusion algorithm is effective.