Image recognition method and system based on brain-like computing platform

Abstract

The invention relates to an image recognition method and system based on a brain-like computing platform. The method comprises the following steps: constructing a spiking neural network, wherein the spiking neural network comprises an approximate maximum pooling layer, the approximate maximum pooling layer comprises a pre-synaptic neuron cluster and a post-synaptic neuron cluster, the pre-synaptic neuron cluster comprises a plurality of neuron groups which are arranged in sequence, the number of pre-synaptic neurons in each neuron group is the same, the pre-synaptic neurons in the ith neuron group in the pre-synaptic neuron cluster are all connected with the ith post-synaptic neuron in the post-synaptic neuron cluster, and when spiking input by the pre-synaptic neurons in the ith neuron group exceeds a set threshold value in an accumulated manner, the ith post-synaptic neuron gives out spiking once; deploying the trained spiking neural network to a brain-like computing platform; and carrying out image recognition by adopting the spiking neural network deployed on a brain-like computing platform. According to the invention, the energy consumption of image recognition during hardware platform calculation is reduced.

Description

technical field [0001] The invention relates to the technical field of neural networks, in particular to an image recognition method and system based on a brain-inspired computing platform. Background technique [0002] Using convolutional neural networks to perform image recognition on commonly used hardware platforms such as CPUs often requires high energy consumption. Using spiking neural networks to complete image recognition tasks on brain-inspired computing platforms can effectively reduce power consumption and achieve similar accuracy to convolutional neural networks on CPU and other hardware platforms. quantity. [0003] In the spiking neural network, the maximum pooling performed by simulating the convolutional neural network has not yet been truly realized on the brain-inspired computing platform, and most of the maximum pooling operations are performed through simulator simulation. [0004] In the max pooling of the spiking neural network implemented by the simu...

AI Technical Summary

Problems solved by technology

However, this method is difficult to implement on the hardware platform, and consumes memory, which is not conducive to the calculation of neurons.

Images