FPGA-based Yolov3 network computing acceleration system and acceleration method thereof

Abstract

The invention relates to a Yolov3 network computing acceleration system based on an FPGA and an acceleration method thereof. The system comprises an ARM and FPGA platform architecture, an off-chip storage area, an AXI_M interface and an AXI_S interface, the ARM platform architecture comprises a core processor and a data and memory controller, and the FPGA platform architecture comprises an acceleration core unit, an input cache end and an output cache end; and the core processor comprises an ARM Cortex-A53 CPU and an L2 cache region, the off-chip storage region comprises an SD card and an external DDR4, and the acceleration core unit comprises a data matrix vector array and a calculation module. The input cache end and the output cache end adopt a multi-channel parallel reading and writing-back mode to replace a traditional single-channel reading and writing mode, and the bandwidth of the Zynq chip is utilized to the maximum extent. Double cache regions and a register array are designed at the input cache end, efficient data multiplexing is achieved, and the bandwidth is multiplied.

Description

technical field [0001] The invention relates to the technical field of Yolov3 network computing acceleration, and relates to an FPGA-based Yolov3 network computing acceleration system and an acceleration method thereof. Background technique [0002] As a new generation of computing model, deep learning has played an important role in many fields in recent years. Deep learning algorithms represented by convolutional neural networks have made major breakthroughs in the field of computer vision, such as image classification, object detection and other specific applications. With the advent of the Internet of Things era and the rise of edge computing, the development trend of the future society is the interconnection of everything and the perception of everything. Therefore, it is of great practical significance to deploy deep learning algorithms on the embedded side. [0003] Deep learning algorithms are characterized by intensive computing and storage, and consume a lot of co...

AI Technical Summary

Problems solved by technology

However, YOLOv3 has a large amount of calculation and is difficult to deploy on the embedded side, and the current hardware accelerator lacks a system design solution for YOLOv3 network acceleration

Images