SOC-based data reuse convolutional neural network accelerator

Abstract

The invention provides an SOC-based data reuse convolutional neural network accelerator. Input data of image input, weight parameters, offset parameters and the like of a convolutional neural networkis grouped; a large amount of the input data is classified into reusable block data; and reusable data blocks are read through a control state machine. The convolutional neural network has large parameter quantity and requires strong calculation capability, so that the convolutional neural network accelerator needs to provide very large data bandwidth and calculation capability. A large load is subjected to reusable segmentation, and data reuse is realized through a control unit and an address generation unit, so that the calculation delay and the required bandwidth of the convolutional neuralnetwork are shortened and reduced, and the calculation efficiency is improved.

Description

technical field [0001] The invention relates to a SOC-based data multiplexing convolutional neural network accelerator, in particular to the convolutional neural network for embedded devices, and belongs to the field of embedded applications. Background technique [0002] With the continuous development and optimization of convolutional neural network CNN (Convolutional Neural Network), it has been widely used in the field of pattern recognition, including image recognition, target recognition, image segmentation, target tracking and other fields, and has achieved remarkable results. Shows the dominance of convolutional neural networks in pattern recognition algorithms. [0003] However, deep convolutional neural networks consume a lot of computing resources and storage resources, and cannot be directly applied to the embedded side. Convolutional neural network AlexNet for image recognition, convolution and fully connected operations include a total of 1.45G operations, and...

AI Technical Summary

Problems solved by technology

[0004] The problem to be solved by the technology of the present invention is: to overcome the shortcomings of the existing convolutional neural network, such as low operational efficiency and the time delay and waste of power consumption caused by a large number of accesses to external memory, and to provide a data multiplexing convolution based on SOC Neural network accelerator, making full use of the reuse characteristics of input data and convolution kernel data, improves the computing performance of embedded devices for convolutional neural networks

Images