[1]于 翔,周 波.基于多模态融合的室内人体跟踪技术研究[J].计算机技术与发展,2023,33(02):38-43.[doi:10. 3969 / j. issn. 1673-629X. 2023. 02. 006]
　YU Xiang,ZHOU Bo.Research on Indoor Human Tracking Technology Based on Multi-modal Fusion[J].,2023,33(02):38-43.[doi:10. 3969 / j. issn. 1673-629X. 2023. 02. 006]
点击复制

基于多模态融合的室内人体跟踪技术研究()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
33
期数:
2023年02期
页码:
38-43
栏目:
媒体计算
出版日期:
2023-02-10

文章信息/Info

Title:
Research on Indoor Human Tracking Technology Based on Multi-modal Fusion
文章编号:
1673-629X(2023)02-0038-06
作者:
于 翔; 周 波
东南大学 自动化学院,江苏 南京 210096
Author(s):
YU Xiang; ZHOU Bo
School of Automation,Southeast University,Nanjing 210096,China
关键词:
移动机器人; 多模态融合; 室内人体跟踪; 人体检测; 数据关联
Keywords:
mobile robots; multi-modal fusion; indoor human tracking; human detection; data association
分类号:
TP391. 4
DOI:
10. 3969 / j. issn. 1673-629X. 2023. 02. 006
摘要:
传统的室内人体跟踪一般基于单独的相机或者二维激光雷达进行,基于二维点云的人腿检测在面对桌 / 椅腿较多的环境时准确度明显不足,而基于图像的人体检测在室内光线较暗时也会出现鲁棒性不足的问题,这些单模态的方法往往难以在干扰嘈杂的环境工作。 因此,该文提出了一种基于多模态融合的室内人体跟踪方法,训练一个 Adaboost 分类器将点云段分为人腿 / 非人腿,使用成熟的 YOLOv3 网络对图像中的人体进行检测,结合点云段分类结果和视觉人体边界框完成基于概率数据关联的模态融合,最后使用扩展卡尔曼滤波器完成人体跟踪。 对该方法在嘈杂的实验室环境下和晦暗的走廊环境下进行测试,结果表明多模态融合的室内人体检测取得了优于单模态的效果,验证了该系统的鲁棒性和有效性,适用于移动机器人的室内人体跟踪任务。
Abstract:
Traditional indoor human tracking is usually based on a separate camera or two-dimensional LiDAR. The accuracy of two-dimensional point clouds- based human leg detection is significantly inadequate when facing the environments having more table / chairlegs. Image-based human detection also has the problem of insufficient robustness when the light is dim. These single-modal methodsare often difficult to work in noisy environments. Therefore,we present an indoor human tracking method based on multi-modal fusion.An Adaboost classifier is trained to divide the segments of point clouds into human legs and non - human legs. The mature YOLOv3network is used to detect human body in the image. The classification results of segments and the visual human bounding boxes arecombined to complete the modal fusion based on probability data association. Finally,the extended Kalman filter is used to completehuman tracking. The experiments in noisy laboratory environment and dark corridor environment show that the proposed algorithmachieves better results than three single-modal methods. The robustness and effectiveness of the proposed system are verified and it issuitable for indoor human tracking tasks of mobile robots.

相似文献/References:

[1]王娜 马昕.基于细化算法的移动机器人拓扑地图创建[J].计算机技术与发展,2009,(10):11.
　WANG Na,MA Xin.Mobile Robot Topological Map Building Based on Thinning Algorithm[J].,2009,(02):11.
[2]范莉丽 王奇志.改进的生物激励神经网络的机器人路径规划[J].计算机技术与发展,2006,(04):19.
　FAN Li-li,WANG Qi-zhi.Robot Path Planning of Modified Biologically Inspired Neural Networks[J].,2006,(02):19.
[3]王肖青 王奇志.传统人工势场的改进[J].计算机技术与发展,2006,(04):96.
　WANG Xiao-qing,WANG Qi-zhi.An Evolutionary Method of Traditional Artificial Potential Field[J].,2006,(02):96.
[4]吕凌 曾碧.基于评估和分工合作并行蚁群机器人路径规划[J].计算机技术与发展,2011,(09):10.
　Lü Ling,ZENG Bi.Path Planning for Robot Introduction Parallel Ant Colony Algorithm Based on Division of Labor and Assessment[J].,2011,(02):10.
[5]张璐 张国良 张维平 敬斌.基于粒子群三次样条优化的局部路径规划方法[J].计算机技术与发展,2012,(11):145.
　ZHANG Lu,ZHANG Guo-liang,ZHANG Wei-ping,et al.Local Path Planning Algorithm Based on Particle Swarm Optimization of Cubic Splines[J].,2012,(02):145.
[6]徐丁,朱擎飞,叶晓东.遗传算法在移动机器人路径规划中的应用[J].计算机技术与发展,2013,(11):112.
　XU Ding,ZHU Qing-fei,YE Xiao-dong.Application of Genetic Algorithm in Mobile Robot Path Planning[J].,2013,(02):112.
[7]邵曦,陶凯云. 基于音乐内容和歌词的音乐情感分类研究[J].计算机技术与发展,2015,25(08):184.
　SHAO Xi,TAO Kai-yun. Research on Music Emotion Classification Based on Music Content and Lyrics[J].,2015,25(02):184.
[8]张俊溪,米国际,王鑫,等.基于进化算法和模糊控制的机器人路径规划[J].计算机技术与发展,2018,28(06):49.[doi:10.3969/ j. issn.1673-629X.2018.06.011]
　ZHANG Jun-xi,MI Guo-ji,WANG Xin,et al.Research on Path Planning of Robot Based on Evolutionary Algorithm and Fuzzy Control Algorithm[J].,2018,28(02):49.[doi:10.3969/ j. issn.1673-629X.2018.06.011]
[9]李登峰,杨 曦.基于改进智能水滴算法的移动机器人路径规划[J].计算机技术与发展,2019,29(12):49.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 009]
　LI Deng-feng,YANG Xi.Path Planning of Mobile Robot Based on Improved Intelligent Water Drop Algorithm[J].,2019,29(02):49.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 009]
[10]何 松,孙 静,郭乐江,等.基于激光 SLAM 和深度学习的语义地图构建[J].计算机技术与发展,2020,30(09):88.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 016]
　HE Song,SUN Jing,GUO Le-jiang,et al.Semantic Mapping Based on Laser SLAM and Deep Learning[J].,2020,30(02):88.[doi:10. 3969 / j. issn. 1673-629X. 2020. 09. 016]

更新日期/Last Update: 2023-02-10