SOPC-based Implementation of Convolutional Neural Network

Abstract

Convolutional Neural Network (CNN) is an algorithm widely used in the field of deep learning. Due to the large number of intensive parallel data operations, the use of CPU to implement the CNN serially consumes too much time. In view of the above research background, the System-on-a-Programmable-Chip (SOPC) implementation and acceleration modules of CNN are designed by using the Zynq-7035 development platform launched by Xilinx as the experimental platform. The experiment results show that our method is effective and efficiency for the target image classification on the development platform.