首页 > 论文 > 光子学报 > 48卷 > 7期(pp:701002--1)

用于弹载线阵激光雷达的卷积神经网络目标识别

Convolutional Neural Network Target Recognition for Missile-borne Linear Array LiDAR

  • 摘要
  • 论文信息
  • 参考文献
  • 被引情况
  • PDF全文
分享:

摘要

为了提高末敏弹在复杂背景条件下对装甲目标的识别能力, 将线阵激光雷达作为探测器, 结合卷积神经网络对线阵激光雷达距离像进行目标分类与识别.利用末敏弹边旋转边下降的运动特点, 实现对扫描区域的距离成像, 并通过采样率控制及插值等算法将原始距离像构造成适用于卷积神经网络的灰度像.针对弹载高实时性、小体积和低功耗的要求, 建立了由两层卷积层和一层全链接层构成的浅层卷积网络, 选用Xilinx ZYNQSoC 芯片作为硬件平台, 通过基于HLS技术和SDSoC开发环境将卷积操作放在端进行硬件并行加速.缩比模拟试验结果验证了该方法具有较高的目标识别精度, 对复杂背景下的装甲目标也能有效识别.ZYNQSoC的PL硬件相较于普通CPU方案, 加速性能提升了5倍, 能够满足弹载的要求.

Abstract

In order to improve the detection ability of the terminal sensitive projectile to the armored target under complex background conditions, the linear array LiDAR is used as the detector and the convolutional neural network is combined to classify and identify the range profile of the linear array LiDAR. The range imaging of the scanning area is realized by using the steady-state motion characteristics of the terminal sensitive projectile dropping while rotating. The original range profile is constructed into a range profile suitable for convolutional neural network by sampling rate control and interpolation. The convolutional neural network consisting of two convolutional layers and one full link layer are established to meet the requirements of high real-time, small size and low power consumption on missile-borne. Xilinx ZYNQSoC chip is selected as hardware platform, and hardware acceleration is achieved by placing convolution operation on PL end of ZYNQSoC based on HLS technology and SDSoC development environment. The scaled simulation experiment proves that the method has high target recognition accuracy and can effectively recognize armored targets in complex background. Through placing convolution operation on PL end of ZYNQSoC, the performance acceleration is five times faster than that of CPU, which can meet the requirements of missile-borne.

广告组1 - 空间光调制器+DMD
补充资料

中图分类号:TN958

DOI:10.3788/gzxb20194807.0701002

基金项目:陆军重点预研项目(No.301070201)

收稿日期:2019-02-18

修改稿日期:2019-04-29

网络出版日期:--

作者单位    点击查看

武军安:南京理工大学 智能弹药技术国防重点学科实验室, 南京 210094
郭锐:南京理工大学 智能弹药技术国防重点学科实验室, 南京 210094
刘荣忠:南京理工大学 智能弹药技术国防重点学科实验室, 南京 210094
柯尊贵:西南技术物理研究所, 成都 610041

联系人作者:武军安(574732664@qq.com)

备注:武军安(1989-), 男, 博士研究生, 主要研究方向为末敏弹的目标探测识别技术.

【1】SU Po, RONG Yong-jie, XU Jian-sheng, et al. Research on the recognition and position method for linear array laser radar of terminal sensitive projectile[J]. Journal of Projectiles, Rockets, Missiles and Guidance, 2017, 37(1): 161-164.
苏坡,戎永杰,许建胜,等. 末敏弹线阵激光雷达识别与定位方法研究[J]. 弹箭与制导学报,2017, 37(1): 161-164.

【2】WU Jun-an, GUO Rui, LIU Rong-zhong, et al. Armored target extraction based on linear array LiDAR of terminal sensitive sub-ammunition[J]. Infrared and Laser Engineering, 2018, 48(3): 0330002.
武军安,郭锐,刘荣忠,等. 末敏弹线阵激光雷达对地面装甲目标的提取方法[J].红外与激光工程,2018, 48(3): 0330002.

【3】MA Chao-jie, YANG Hua, LI Xiao-xia, et al. Implementation of automatic target recognition by imagine Ladar in complex scenes[J]. Optics and Precision Engineering, 2009, 17(7): 1714-1720
马超,杨华,李小霞,等. 复杂场景下应用成像Ladar 的自动目标识别[J]. 光学精密工程, 2009,17(7):1714-1720

【4】HUANG Tao, HU Yi-hua, ZHAO Gang. Target extraction and classification base on imaging LADAR range image[J]. Journal of Infrared and Millimeter Waves, 2011, 30(2): 170-183.
黄涛,胡以华,赵钢.基于激光成像雷达距离图像的目标提取与分类技术[J]. 红外与毫米波学报, 2011,30(2):179-183.

【5】HANS C P, TRYM V H, HALVOR A, et al. Extraction and classification of vehicles in Ladar imagery[C]. SPIE, 2013, 8731(4): 507-522.

【6】RUSU R B, BRADSKI G, THIBAUX R, et al. Fast 3D recognition and pose using the viewpoint feature histogram[C]. IEEE International Conference on Intelligent Robots and Systems, 2010: 2155-2162.

【7】ALDOMA A, VINCZE M, BLODOW N, et al. CAD-model recognition and 6DOF pose estimation using 3D cues[C]. IEEE International Conference on Computer Vision Workshops, 2011: 585-592.

【8】RUSU R B, BLODOW N, BEETZ M. Fast point feature histograms (FPFH) for 3D registration[C]. IEEE International Conference on Robotics and Automation, 2009: 1848-1853.

【9】PRAKHYA S M, LIU B, LIN W, et al. B-SHOT:a binary 3D feature descriptor for fast key point matching on 3D point clouds[J]. Autonomous Robots, 2017, 41(7): 1501-1520.

【10】LECUN Y, BENGIO Y, HINTION G. DEEP learning[J]. Nature, 2015, 521(7553): 436-444.

【11】HE K, ZHANG X, REN S, et al. Deep residual learning for image recognition[J]. Computer Science, 2015. arXiv:1512.033385v1.

【12】HE X, WANG A L, GHAMISI P, et al. LiDAR data classification using spatial transformation and CNN[J]. IEEE Geoscience and Remote Sensing Letters, 2018, (99):1-5.

【13】YANG W H, YU F. LiDAR image classification based on convolutional neural networks[C]. International Conference on Computer Network, Electronic and Automation, 2017, 37: 221-225.

【14】CHEN Y S, LI C Y, GHAMISI P, et al. Deep fusion of hyperspectral and LiDAR data for thematic classification[C]. IEEE International Geoscience And Remote Sensing Symposium, 2016, 3590-3504.

【15】GAO H B, CHENG B, WANG J Q, et al. Object classification using CNN-based fusion of vision and LiDAR in autonomous vehicle environment[J]. IEEE Transactions on Industrial Informatics, 2018, (99): 4224-4231.

【16】ZHOU Y, TUZEL O. Voxelnet: End-to-end learning for point cloud based 3d object detection[J]. 2017, Arxiv: 1711.06396

【17】STRIGL D, KOFLER K, PODLIPNIG S. Performance and scalability of GPU-based convolutional neural networks[C]. Proceedings of Euromicro International Conference on Parallel, Distributed and Network-Based Processing. Piscataway, NJ: IEEE Press, 2010: 317-324.

【18】LIU Zhi-cheng, ZHU Yong-xin, WANG Hui, et al. Parallel acceleration design of convolutional neural network based on FPGA[J]. Microelectronics & Computer, 2018, 35(10): 80-85.
刘志成,祝永新,汪辉,等.基于FPGA的卷积神经网络并行加速结构设计[J]. 微电子学与计算机. 2018, 35(10): 80-85.

【19】CAI Rui-chu, YU Yang, ZHONG Chun-rong, et al. Design and scheduling of convolutional neural network accelerator for cloud FPGAs[J]. Application Research of Computers. DOI: 10.19734/j.issn.1001-3695.2018.05.0507
蔡瑞初,余洋,钟椿荣,等.面向云端FPGA的卷积神经网络加速器的设计及其调度[J]. 计算机应用研究. DOI: 0.19734/j.issn.1001-3695.2018.05.0507

【20】Xilinx. 2017. Xilinx Zynq-7000 All Programmable SoC[EB/OL]. [2019-04-18]. https://www.xilinx.com/products/silicon-devices/soc/zynq-7000.html.

引用该论文

WU Jun-an,GUO Rui,LIU Rong-zhong,KE Zun-gui. Convolutional Neural Network Target Recognition for Missile-borne Linear Array LiDAR[J]. ACTA PHOTONICA SINICA, 2019, 48(7): 0701002

武军安,郭锐,刘荣忠,柯尊贵. 用于弹载线阵激光雷达的卷积神经网络目标识别[J]. 光子学报, 2019, 48(7): 0701002

您的浏览器不支持PDF插件,请使用最新的(Chrome/Fire Fox等)浏览器.或者您还可以点击此处下载该论文PDF