[1]朱齐丹,韩瑜,蔡成涛.全景视觉非线性核相关滤波目标跟踪技术[J].哈尔滨工程大学学报,2018,39(07):1220-1226.[doi:10.11990/jheu.201705042]
 ZHU Qidan,HAN Yu,CAI Chengtao.Omni-directional visual object tracking using nonlinear kernelized correlation filters[J].hebgcdxxb,2018,39(07):1220-1226.[doi:10.11990/jheu.201705042]
点击复制

全景视觉非线性核相关滤波目标跟踪技术(/HTML)
分享到:

《哈尔滨工程大学学报》[ISSN:1006-6977/CN:61-1281/TN]

卷:
39
期数:
2018年07期
页码:
1220-1226
栏目:
出版日期:
2018-07-05

文章信息/Info

Title:
Omni-directional visual object tracking using nonlinear kernelized correlation filters
作者:
朱齐丹 韩瑜 蔡成涛
哈尔滨工程大学 自动化学院, 黑龙江 哈尔滨 150001
Author(s):
ZHU Qidan HAN Yu CAI Chengtao
College of Automation, Harbin Engineering University, Harbin 150001, China
关键词:
全景视觉目标跟踪非线性核相关滤波器旋转不变性HOG
分类号:
TP242.6
DOI:
10.11990/jheu.201705042
文献标志码:
A
摘要:
为了实现全景视觉下目标跟踪,本文提出了基于非线性核相关滤波的全景视觉目标跟踪算法,该算法将岭回归与循环样本矩阵和经典的相关滤波联系在一起。通过核技巧,建立了非线性回归函数,解决了全景成像畸变的影响。通过在极坐标系下构造傅里叶形式的梯度方向直方图特征描述子(HOG),保证了全景目标的旋转不变性。针对全景成像特点设计自适应机制和基于极坐标表示的目标搜索机制。以OTB100库中存在平面内旋转情况的图像序列与原方法进行对比实验,本文算法跟踪精度提高13.5%。并在全景图像上进行实验,相比于原方法,本文算法能够实现有效全景视觉目标跟踪。

参考文献/References:

[1] BONIN F, ORTIZ A, OLIVER G. Visual navigation for mobile robots:a survey[J]. Journal of intelligent and robotic systems, 2008, 53(3):263-296.
[2] CHEN Chusong, HSIEH W T, CHEN J H. Panoramic appearance-based recognition of video contents using matching graphs[J]. IEEE transactions on systems, man, and cybernetics, Part B (Cybernetics), 2004, 34(1):179-199.
[3] 朱齐丹, 徐从营, 蔡成涛, 等. 全景图像海天线提取及舰船目标自动检测[J]. 计算机测量与控制, 2014, 22(8):2350-2352.ZHU Qidan, XU Congying, CAI Chengtao, et al. Sea-sky line extraction and ship target automatic detection of panoramic image[J]. Computer measurement & control, 2014, 22(8):2350-2352.
[4] 谢洪乐, 朱齐丹, 刘鹏. 基于极大似然估计的全景立体视觉机器人自定位方法[J]. 应用科技, 2017, 44(5):40-45, 51.XIE Hongle, ZHU Qidan, LIU Peng. Self-localization of a robot with omnidirectional stereoscopic vision based on maximum likelihood estimation algorithm[J]. Applied science and technology, 2017, 44(5):40-45, 51.
[5] SMEULDERS A W M, CHU D M, CUCCHIARA R, et al. Visual tracking:an experimental survey[J]. IEEE transactions on pattern analysis and machine intelligence, 2014, 36(7):1442-1468.
[6] FURGAL P, BARFOOT T D. Visual teach and repeat for long-range rover autonomy[J]. Journal of field robotics, 2010, 27(5):534-560.
[7] LI Xi, HU Weiming, SHEN Chunhua, et al. A survey of appearance models in visual object tracking[J]. ACM transactions on intelligent systems and technology, 2013, 4(4):58.
[8] HENRIQUES J F, CASEIRO R, MARTINS P, et al. High-speed tracking with kernelized correlation filters[J]. IEEE transactions on pattern analysis and machine intelligence, 2015, 37(3):583-596.
[9] DALAL N, TRIGGS B. Histograms of oriented gradients for human detection[C]//Proceedings of 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. San Diego, CA, USA, 2005:886-893.
[10] LIU Kun, SKIBBE H, SCHMIDT T, et al. Rotation-invariant HOG descriptors using fourier analysis in polar and spherical coordinates[J]. International journal of computer vision, 2014, 106(3):342-364.
[11] WU Yi, LIM J, YANG M H. Object tracking benchmark[J]. IEEE transactions on pattern analysis and machine intelligence, 2015, 37(9):1834-1848.

相似文献/References:

[1]陆军,孙凌丽,穆海军,等.基于全景视觉的机器人相互定位的研究[J].哈尔滨工程大学学报,2010,(04):0.
[2]庄佳园,徐玉如,万磊,等.基于雷达图像的水面无人艇目标检测技术[J].哈尔滨工程大学学报,2012,(02):129.[doi:10.3969/j.issn.1006-7043.201012029]
 ZHUANG Jiayuan,XU Yuru,WAN Lei,et al.Target detection of an unmanned surface vehicle based on a radar image[J].hebgcdxxb,2012,(07):129.[doi:10.3969/j.issn.1006-7043.201012029]
[3]张龙,崔乃刚,杨峰,等.高阶容积卡尔曼滤波及其在目标跟踪中的应用[J].哈尔滨工程大学学报,2016,37(04):573.[doi:10.11990/jheu.201412079]
 ZHANG Long,CUI Naigang,YANG Feng,et al.High-degree cubature Kalman filter and its application in target tracking[J].hebgcdxxb,2016,37(07):573.[doi:10.11990/jheu.201412079]

备注/Memo

备注/Memo:
收稿日期:2017-05-09。
基金项目:国家自然科学基金项目(61673129);黑龙江省自然科学基金项目(F201414);中央高校基本科研业务费专项资金项目(HEUCF160418).
作者简介:朱齐丹(1963-),男,教授,博士生导师;韩瑜(1991-),男,博士研究生.
通讯作者:韩瑜,E-mail:hanyu@hrbeu.edu.cn
更新日期/Last Update: 2018-07-07