[1]佟璐,申萍,王立德,等.基于深度图像的列车乘客目标识别技术研究[J].机车电传动,2014,(02):46-49.[doi:10.13890/j.issn.1000-128x.2014.02.031]
 TONG Lu,SHEN Ping,WANG Li-de,et al.Human Recognition Based on the Depth Image in Rail Vehicle[J].Electric Drive for Locomotives,2014,(02):46-49.[doi:10.13890/j.issn.1000-128x.2014.02.031]
点击复制

基于深度图像的列车乘客目标识别技术研究()
分享到:

机车电传动[ISSN:1000-128X/CN:43-1125/U]

卷:
期数:
2014年02期
页码:
46-49
栏目:
研究开发
出版日期:
2014-03-10

文章信息/Info

Title:
Human Recognition Based on the Depth Image in Rail Vehicle
文章编号:
1000-128X(2014)02-0046-04
作者:
佟璐1申萍1王立德1刘明坤2
1. 北京交通大学 2. 南车青岛四方机车车辆股份有限公司
Author(s):
TONG Lu1 SHEN Ping1 WANG Li-de1 LIU Ming-kun2
(1. School of Electrical Engineering, Beijing Jiaotong University, Beijing 100044,China;2. Technology Center, CSR Qingdao Sifang Locomotive & Rolling Stock Co., Ltd., Qingdao, Shandong 266111, China)
关键词:
深度图像法向量RGB-D目标识别特征提取客流轨道交通
Keywords:
depth image normal vector RGB-D object recognition feature extraction passenger flow rail transit
分类号:
U298;TP277
DOI:
10.13890/j.issn.1000-128x.2014.02.031
文献标志码:
A
摘要:
针对轨道交通中高密度客流的特点,提出了一种基于深度图像的局部法向量特征提取方法,利用球面坐标系参数( , )表示物体表面切平面法向量的特征信息,通过提取检测区域的局部法向量作为目标特征,以实现目标识别及头部轮廓提取。采用深度图像能有效地解决轨道车辆高度及光照等环境因素的局限性,并通过试验验证该方法的实时性和准确性。
Abstract:
A method based on depth image local normal vector characteristic, described by spherical coordinate parameters ( , ), was proposed in this paper and suited to rail vehicle with the high-density passengers. This method implemented the human recognition by capturing the local normal vector, and then extracting the head contour. Depth image was a better way to solve the limitations of environmental factors effectively, such as height and illumination. Finally, some experiments were taken to test and verify the real-time performance and accuracy.

参考文献/References:

[1] 赵军伟. 基于数学形态学和HIS颜色空间的人头检测[J]. 山东大学学报:工学版,2013,43(2) :6-10.

[2] Ganapathi V, Plagemann C. Real Time Motion Capture Usinga Single Time-Of-Flight Camera[C]. San Francisco:2010 IEEEConference on Computer Vision and Pattern Recognition(CVPR), 2010:755-762.
[3] Ikemura S, Fujiyoshi H. Real-Time Human Detection usingRelational Depth Similarity Features[C]. Queenstown:ComputerVision-ACCV 2010. 10th Asian Conference on Computer Vision,2011:25-38.
[4] Jain, Himanshu Prakash,Subramanian, et al. Real-time UpperbodyHuman Pose Estimation using a Depth Camera[C]. Rocquencourt:Computer Vision/Computer Graphics CollaborationTechniques - 5th International Conference, MIRAGE 2011,2010:227-238.[5] 李红波,丁林建,冉光勇. 基于Kinect深度图像的人体识别分析[J]. 数字通信,2012(4) :21-26.
[6] 蔡睿妍. 基于Kinect的多点触控系统的设计与实现[J]. 电脑知识与技术,2012(16) :3987-3989.
[7] 林鹏,张超,李竹良,等. 基于深度图像学习的人体部位识别[J]. 计算机工程,2012(16) :185-188.
[8] Tang Shuai, Wang Xiaoyu. Histogram of Oriented NormalVectors for Object Recognition with a Depth Sensor[C]. Daejeon:Computer Vision-ACCV 2012. 11th Asian Conference onComputer Vision, 2013:525-538.
[9] Padeleris P, Zabulis X. Head pose estimation on depth data basedon particle swarm optimization[C]. Providence:2012 IEEEComputer Society Conference on Computer Vision and PatternRecognition Workshops (CVPR Workshops), 2012:42-49.
[10]Rusi Antonov Filipov. Pylon grid: A fast method for humanhead detection in range images[J]. Neurocomputing, 2013,100 :74-85.

备注/Memo

备注/Memo:
作者简介:佟璐(1990-),女,硕士研究生,主要从事图像与视频处理、控制网络及多媒体数据传输技术的研究。
更新日期/Last Update: 2014-03-10