Spiking Neural Network and Image Processing

In today's world image recognition by the computer is the most concern issue. Recognition of handwriting has become very useful to make the machine learn to interact between human and computer . For pattern classification and recognition ,Neural Networks are the most powerful tools. This book depicts the strategies for hand written English characters (upper and lowercase) .We focus on classification and recognition based on Leaky-Integrate-and fire neuron model and Izhikevich neuron model which are Spiking Neural Network models.This book illustrates about the simulation of Leaky integrate and fire neuron model and Izhikevich neuron model ,and the results are compared in terms of accuracy, simulation time and firing rates for analysis with some further improvements.The successful simulation of few SNN models like LIF model and Izhikevich model for implementing an algorithm as a pattern classifier for English alphabet have been discussed . We report on advantages of SNN such as Mathematical model computation, High Accuracy, Low Power,Less Area, Useful for real time applications for further enhancement and improvement in terms of cost effectiveness.

Autorentext

Soni Adit Chaturvedi è Ph.D[ Ingegneria Elettronica] M.Tech (Elettronica) e B.E. (E &C) Attualmente lavora come professore associato e capo in E &C Engg. Dipartimento, PIET, Nagpur. Ha un'esperienza totale di insegnamento di 25 anni, ha un capitolo di libro in Springer book e 20 contributi di ricerca in riviste internazionali repitate, Area: ANN, DIP, DSD.