本文已被:浏览 10次 下载 1次
中文摘要: 在红外目标识别领域,基于卷积神经网络的深度学习算法的识别精度已远远超过了传统模式识别算法,但神经网络的实现需要庞大的计算和存储,难以在无人机等嵌入式平台上进行部署。针对此问题,将通道级量化策略和梯度的近似优化训练引入到了低比特神经网络模型的建立中,并提出了一种可充分利用硬件计算资源的FPGA加速器,其整体平均性能为65.6GOPS。与其他相关工作的对比表明,低比特量化方法及其FPGA加速器实现,可以为嵌入式红外目标识别系统提供一种能效高、识别精度高的解决方案。
中文关键词: 红外目标识别;卷积神经网络;FPGA加速器;高能效
Abstract:In the field of infrared object detection, the accuracy of deep learning algorithm based on convolutional neural network has exceeded that of traditional pattern recognition algorithms. However, the implementation of neural networks requires huge computing and storage demands, and it is difficult to deploy them on embedded platforms such as drones. To overcome the above problem, this paper introduces a channel-wise quantization strategy and the approximate optimization of the gradient into the training of low-bit neural network model. Besides, a hardware computing resources utilized FPGA accelerator is proposed to implement the model, whose overall performance is 65.6 GOPS. Comparisons with other related works show that the low-bit quantization method and the proposed FPGA accelerator implementation can provide a solution with high energy efficiency and high recognition accuracy for the embedded infrared object detection system.
keywords: infrared object detection;convolutional neural network;FPGA accelerator;high energy efficiency
文章编号:20200606 中图分类号:TN215; TN216 文献标志码:A
基金项目:上海航天技术研究院-上海交大航天先进技术联合研究中心资助项目(USCAST2019-24)
作者 | 单位 |
黄家明 | 上海交通大学 电子信息与电气工程学院 |
陈寰 | 上海航天控制技术研究所 |
史庆杰 | 上海航天控制技术研究所 |
陈海宝 | 上海交通大学 电子信息与电气工程学院 |
Author Name | Affiliation |
HUANG Jiaming | School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University |
CHEN Huan | Shanghai Aerospace Control Technology Institute |
SHI Qingjie | Shanghai Aerospace Control Technology Institute |
CHEN Haibao | School of Electronic Information and Electrical Engineering, Shanghai Jiao Tong University |
引用文本:
黄家明,陈寰,史庆杰,陈海宝.基于FPGA的红外目标识别神经网络加速器设计[J].飞控与探测,2020,(6):66-75.
黄家明,陈寰,史庆杰,陈海宝.基于FPGA的红外目标识别神经网络加速器设计[J].飞控与探测,2020,(6):66-75.