Abstract
Owing to the global coupling and pulse synchronization characteristic, pulse coupled neural networks(PCNN) has been proved to be suitable for image fusion. In this paper, a new image fusion algorithm based on adaptive PCNN is presented. Compared with the traditional algorithms where the linking strength of each neuron is assigned the same value, this algorithm uses the features of each pixel, e.g. energy of Laplacian and standard deviation, as its value, so that the linking strength of each pixel can be chosen adaptively. Experimental results demonstrate that the proposed algorithm outperforms Laplacian-based, wavelet-based and PCNN-based fusion algorithms.
Original language | English |
---|---|
Pages (from-to) | 779-782 |
Number of pages | 4 |
Journal | Guangdianzi Jiguang/Journal of Optoelectronics Laser |
Volume | 21 |
Issue number | 5 |
State | Published - May 2010 |
Keywords
- Energy of Laplacian(EOL)
- Features
- Image fusion
- Linking strength
- Pulse coupled neural networks(PCNN)
- Standard deviation(SD)