Digit Recognition in Spiking Neural Networks using Wavelet Transform

Nowadays, given the rapid progress in pattern recognition, new ideas such as theoretical mathematics can be exploited to improve the efficiency of these tasks. In this paper, the Discrete Wavelet Transform (DWT) is used as a mathematical framework to demonstrate hand-written digit recognition in spiking neural networks (SNNs). The motivation behind this method is that the wavelet transform can divide the spike information and noise into separate frequency sub-bands, and also store the time information. The simulation results show that DWT is an effective and worthy choice, and brings the network to an efficiency comparable to previous networks in the spiking field. Initially, DWT is applied to MNIST images in the network input. Subsequently, a type of time encoding called constant-current-Leaky Integrate and Fire (LIF) encoding is applied to the transformed data. Following this, the encoded images are input to the multi-layer convolutional spiking network. In this architecture, various wavelets are investigated, and the highest classification accuracy of 99.25% is achieved.