Circle detection using a spiking neural network

The receptive field of neurons plays various roles in biological neural networks. In this paper a spiking neural network model is proposed using a mechanism inspired by the biological receptive field. The network is composed of multiple layers, and the neurons are connected by excitatory and inhibitory synapses. When a visual image presents to the network, location and radius of a circle on the visual image can be obtained from firing rates of the neurons from the corresponding layers. The simulation results show that the network can perform circle detection similar to Hough circle detection and calculations are conducted by a parallel mechanism in a biological manner. This model can be used to explain how a spiking neuron-based network to detect circle, and the high speed parallel mechanism in the model can be used in artificial intelligent systems.