[an error occurred while processing this directive] [an error occurred while processing this directive] [an error occurred while processing this directive]
[an error occurred while processing this directive]

弹箭与制导学报 >

2020 , Vol. 40 >Issue 1: 19 - 23

DOI: https://doi.org/10.15892/j.cnki.djzdxb.2020.01.004

基于轻量化神经网络的目标识别跟踪算法研究

  • 曹昭睿 1 ,
  • 白帆 2 ,
  • 刘凤丽 1 ,
  • 郝永平 2
展开
  • 1 沈阳理工大学机械工程学院, 沈阳 110159
  • 2 沈阳理工大学装备工程学院, 沈阳 110159

通讯简介:白帆,副教授,博士,E-mail:snow-wind-001@163.com。

曹昭睿(1993-),男,辽宁沈阳人,博士研究生,研究方向:目标识别算法与光电成像导引系统。

收稿日期: 2018-10-16

  网络出版日期: 2025-05-30

基金资助

国防科技预先研究项目资助

收起

Design of Target Recognizing and Tracking Algorithm Based on Tiny Convolution Neural Network

  • CAO Zhaorui 1 ,
  • BAI Fan 2 ,
  • LIU Fengli 1 ,
  • HAO Yongping 2
Expand
  • 1 School of Mechanical Engineering, Shenyang Ligong University, Shenyang 110159, China
  • 2 School of Equipment Engineering, Shenyang Ligong University, Shenyang 110159, China

Received date: 2018-10-16

  Online published: 2025-05-30

Fold

摘要

为解决传统目标识别算法对于多尺度、可变速目标的识别性能较差与全尺寸卷积神经网络对硬件计算空间消耗较大的问题,利用轻量化的YOLO卷积神经网络对视频首帧进行目标识别,结合KCF目标跟踪算法与感知哈希算法对完成识别的目标进行跟踪与矫正。优化后的算法能够对复杂目标进行实时识别,对于目标自身变化具有较强的自适应能力。能够为同一计算平台下的飞行控制、自主避障、目标测距等后续控制指令提供计算空间。

关键词： 深度学习; 卷积神经网络; YOLO; KCF跟踪算法; 感知哈希算法

本文引用格式

曹昭睿 , 白帆 , 刘凤丽 , 郝永平 . 基于轻量化神经网络的目标识别跟踪算法研究[J]. 弹箭与制导学报, 2020 , 40(1) : 19 -23 . DOI: 10.15892/j.cnki.djzdxb.2020.01.004

Abstract

In order to solve the problem of poor performance of traditional target recognition algorithms for multi-scale and variable-speed targets and the large consumption of hardware computing space by full-scale convolution neural network, a lightweight YOLO convolution neural network is used to recognize the target in the first frame of video. The target recognition is tracked and corrected by combining KCF target tracking algorithm and perceptual hash algorithm. The optimized algorithm can recognize complex targets in real time, and has strong adaptive ability for the change of objects themselves. It can provide computational space for flight control, autonomous obstacle avoidance, target ranging and other follow-up control commands on the same computing platform.

Key words： deep learning; convolution neural network; YOLO; KCF tracking algorithm; perceptual hash algorithm

[an error occurred while processing this directive]

参考文献

原文顺序 | 文献年度倒序 | 文中引用次数倒序
[1]
REN Shaoqing, HE Kaiming, GIRSHICK R, et al. Faster R-CNN:towards real-time object detection with region proposal networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6):1137-1149.

DOI PMID

[2]
LIU W, ANGUELOV D, ERHAN D, et al. SSD:single shot multibox detector[C]// ECCV. Proceedings of 2016 14th European Conference on Computer Vision. [S.l.: s.n.], 2016:21-37.

[3]
王宇宁, 庞智恒, 袁德明. 基于YOLO算法的车辆实时检测[J]. 武汉理工大学学报, 2016, 38(10):41-46.

[4]
高宗, 李少波, 陈济楠, 等. 基于YOLO网络的行人检测方法[J]. 计算机工程, 2018, 44(5):215-219.

DOI

[5]
魏湧明, 全吉成. 侯宇青阳. 基于YOLO v2的无人机航拍图像定位研究[J]. 激光与光电子学进展, 2017, 54(11):101-110.

[6]
温捷文, 战荫伟, 凌伟林, 等. 实时目标检测算法YOLO的批再规范化处理[J]. 计算机应用研究, 2018, 35(10):3179-3185.

[7]
REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once:unified,real-time object detection[C]// IEEE. Proceedings of 2016 IEEE Conference on Computer Vision and Pattern Recognition. [S.l.]: IEEE, 2016: 779-788.

[8]
梁硕, 陈金勇, 吴金亮, 等. 基于KCF框架的长时间视频目标跟踪算法[J]. 无线电通信技术, 2017, 43(2):55-58.

[9]
郑武兴, 王春平, 付强. 改进的KCF红外空中目标跟踪方法[J]. 激光与红外, 2017, 47(12):1553-1558.

[10]
陈佳兴. 基于KCF和视觉注意的多特征动态融合视觉目标跟踪研究[D]. 成都: 西南交通大学, 2017.

[11]
方正涛, 陈临强. 基于图像感知哈希的行人跟踪算法[J]. 电子科技, 2016, 29(11):112-114.

[12]
黄嘉恒, 李晓伟, 陈本辉, 等. 基于哈希的图像相似度算法比较研究[J]. 大理大学学报, 2017, 2(12):32-37.

Options
文章导航

/

〈 〉

[an error occurred while processing this directive]