The Journal of China Universities of Posts and Telecommunications ›› 2020, Vol. 27 ›› Issue (6): 54-72.doi: 10.19682/j.cnki.1005-8885.2021.0046

Previous Articles     Next Articles

Memristor-based multi-channel pulse coupled neural network for image fusion

Liu Jian, Wu Chengmao, Tian Xiaoping   

  • Received:2020-08-17 Revised:2020-11-28 Online:2020-12-31 Published:2020-12-31
  • Contact: liu jian E-mail:13137739560@163.com
  • Supported by:
    The National Natural Science Foundation of China;the Shaanxi Natural Science Foundation of China

Abstract: Image fusion is widely used in computer vision and image analysis. Considering that the traditional image fusion algorithm has a certain limitation in multi-channel image fusion, a memristor-based multi-channel pulse coupled neural network (M-MPCNN) for image fusion is proposed. Based on a dual-channel pulse coupled neural network (D-PCNN), a novel multi-channel pulse coupled neural network (M-PCNN) is firstly constructed in this paper. Then the exponential growth dynamic threshold model is used to improve the pulse generation of pulse coupled neural network, which can not only avoid multiple ignitions effectively, but can also improve operational efficiency and reduce complexity. At the same time, synchronous capture can also enhance image edge, which is more conducive to image fusion. Finally, the threshold and synaptic characteristics of pulse coupled neural networks (PCNNs) can be well realized by using a memristor-based pulse generator. Experimental results show that the proposed algorithm can fuse multi-source images more effectively than existing state-of-the-art fusion algorithms.

Key words: multi-channel, memristor, pulse coupled neural network