Abstract :
Image fusion plays an important role in many fields such as computer vision, medical image, manufacturing, military, and remote sensing so on. Pulse coupled neural network (PCNN) is derived from the synchronous neuronal burst phenomena in the cat visual cortex. So it is very suitable for image processing. Due to some defects of original PCNN for data fusion, we propose a novel PCNN model - dual- channel PCNN for the first time based on original model, which is specialized in image fusion. In order to explain efficiency and validity of our proposed method, we take two medical images for example to explain further the advantages in comparison to other image fusion methods. Better results are obtained with our approach. Our fused image includes more information than others, which show our method is better and efficient one. Meanwhile our method not only fuses multi-source images very well but also enhances the quality of the fused image.
Keywords :
image fusion; neural nets; cat visual cortex; image fusion; pulse coupled neural network; synchronous neuronal burst phenomena; Application software; Biomedical imaging; Brain modeling; Computer aided manufacturing; Computer vision; Image fusion; Image processing; Military computing; Neural networks; Remote sensing; PCNN; image fusion; image processing; quality assessment;
