Memristor-based circuit implementation of pulse-coupled neural network with dynamical threshold generators Academic Article

abstract

• 2018 Elsevier B.V. Pulse-coupled neural network (PCNN) is inspired from the visual cortex of cats. It is superior to the traditional algorithm in the field of digital image processing. Meanwhile, memristor is considered as an important circuit element to implement brain intelligence hardware. In order to realize the memristor-based circuit of PCNN, a threshold generator is designed to dynamically update the memristance under input excitation with the exponential memristor model. Furthermore, the whole circuit of non-simplified pulse-coupled neural network is realized via memristor. Finally, the image processing function is demonstrated by the proposed circuit via simulation.

authors

• Huang, Tingwen

published proceedings

• NEUROCOMPUTING

author list (cited authors)

• Xie, X., Wen, S., Zeng, Z., & Huang, T.

citation count

• 58

complete list of authors

• Xie, Xudong||Wen, Shiping||Zeng, Zhigang||Huang, Tingwen

publication date

• April 2018

publisher

• Elsevier  Publisher

published in

• Neurocomputing  Journal

keywords

• Dynamical Threshold Generator
• Memristor
• Pulse-coupled Neural Network

Digital Object Identifier (DOI)

• 10.1016/j.neucom.2018.01.024

start page

• 10

end page

• 16

volume

• 284

URL

• http://dx.doi.org/10.1016/j.neucom.2018.01.024