基于改进Mean-Shift 算法的红外小目标跟踪

杨一帆; 田雁; 杨帆; 黄彪

留言板

尊敬的读者、作者、审稿人, 关于本刊的投稿、审稿、编辑和出版的任何问题, 您可以本页添加留言。我们将尽快给您答复。谢谢您的支持!

姓名
邮箱
手机号码
标题
留言内容
验证码

基于改进Mean-Shift 算法的红外小目标跟踪

杨一帆, 田雁, 杨帆, 黄彪

文章导航 > 红外与激光工程 > 2014 > 43(7): 2164-2169

杨一帆, 田雁, 杨帆, 黄彪. 基于改进Mean-Shift 算法的红外小目标跟踪[J]. 红外与激光工程, 2014, 43(7): 2164-2169.

引用本文:

杨一帆, 田雁, 杨帆, 黄彪. 基于改进Mean-Shift 算法的红外小目标跟踪[J]. 红外与激光工程, 2014, 43(7): 2164-2169.

Yang Yifan, Tian Yan, Yang Fan, Huang Biao. Tracking of infrared small-target based on improved Mean-Shift algorithm[J]. Infrared and Laser Engineering, 2014, 43(7): 2164-2169.

Citation:

Yang Yifan, Tian Yan, Yang Fan, Huang Biao. Tracking of infrared small-target based on improved Mean-Shift algorithm[J]. Infrared and Laser Engineering, 2014, 43(7): 2164-2169.

基于改进Mean-Shift 算法的红外小目标跟踪

杨一帆^1,2,
田雁¹,
杨帆^2,3,
黄彪^1,2

1.
中国科学院西安光学精密机械研究所,陕西西安 710119;
2.
中国科学院大学,北京 100049;
3.
中国科学院国家授时中心,陕西西安 710600

详细信息

作者简介:
杨一帆（1987-），男，硕士生，主要从事图像目标跟踪方面的研究工作。Email：yangyifan@yeah.net

中图分类号: TP391

Tracking of infrared small-target based on improved Mean-Shift algorithm

1.
Xi'an Institute of Optics and Precision Mechanics Chinese Academy of Sciences,Xi'an 710119,China;
2.
University of Chinese Academy of Sciences,Beijing 100049 China;
3.
National Time Service Center,Chinese Academy of Sciences,Xi'an,710600,China

摘要: 复杂背景下的红外小目标跟踪在目标跟踪领域一直是重要的研究方向。由于小目标体量小、机动性大，而红外图像大多受到严重的背景噪声和热噪声影响，使得针对红外小目标的跟踪大多出错率高，鲁棒性不强。针对红外小目标的跟踪，提出了一种改进的Mean-Shift 算法。结合图像的统计特性，提出了一种自适应非线性算法对图像进行处理；同时融合了图像的梯度直方图对目标进行描述。实验通过对高强度噪声和高遮挡环境下视频目标进行跟踪，比较了传统Mean-Shift 算法和改进后算法的跟踪效果，结果显示文中提出的改进算法不但可以有效地跟踪目标，而且大幅降低了跟踪窗口与目标之间的相对抖动，增强了跟踪算法的鲁棒性。
- Mean-Shift /
- 红外小目标 /
- 自适应 /
- 非线性 /
- 融合
Abstract: Small-target tracking in infrared imagery with a complex background is always an important task in object tracking fields. Small and manoeuvrable objects in complex clutter and highly noised background usually results in serious false alarm in target tarking for low contrast of infrared imagery. An improved Mean-Shift algorithm to handle the influnce of complex background during tracking the smalltarget in infrared imagery was proposed. This work proposed an adaptive nonlinear machine to help Mean-Shift algorithm to get stable histogram of the interesting areas. This machine expanded the imformation of object histogram refer to the mean value of tracking window, as well as reduced the noise part of tracking window refer to the standard deviation of it. At the same time, algorithm fused the histograms of gradient with histogram of gray-value to discribe the target. To validate the effection of the proposed algorithm, the last part conduct a series tracking experiments which choose highly noised and clutered videos as their candidates. The comparison of the tracking results between tradtional Mean-Shift algotithm and improved Mean-Shift algorithm shows that the proposed algorithm has a more accurate tracking effection. Further more the proposed algorithm highly reduced the wobbleing between small-target and tracking window. This indicates that the improved algorithm achieved more robustness.
- Mean-Shift /
- small-target /
- adaptive /
- nonlinear /
- fusion

[1]
[2]	Comaniciu D, Ramesh V, Meer P. Kernel-based object tracking [J]. IEEE Trans Pattern Anal, 2003(5): 564-575.
[3]	Jin Weiqi, Liu Bin, Fan Yongjie, et al. Review on infrared image detail enhancement techniques [J]. Infrared and Laser Engineering, 2012, 40(12): 2521-2527. (in Chinese) 金伟其, 刘斌, 范永杰. 红外图像细节增强技术研究进展 [J]. 红外与激光工程, 2012, 40(12): 2521-2527.
[4]
[5]
[6]	Tang Zhengyuan, Zhao Jiajia, Yang Jie, et al.Infrared target tracking algorithm based on sparse representation model [J]. Infrared and Laser Engineering, 2012, 41 (5): 1398-1395. (in Chinese) 唐峥远, 赵佳佳, 杨杰, 等. 基于稀疏表示模型的红外目标跟踪算法[J]. 红外与激光工程, 2012, 41(5): 1398-1395.
[7]	David R, Jongwoo L. Adaptive probabilistic visual tracking with incremental subspace update [C]//ICSC, 2003, 2: 234-240.
[8]
[9]	Comaniciu D, Ramesh V, Meer P. Real-time tracking of non-rigid objects using Mean Shift [C]//Proc IEEE Coference on Computer Vision and Pattern Recognition, 2000: 142-149.
[10]
[11]
[12]	Robert T. Mean-shift blob tracking through scale space [C]// ECCV, 2004, 3022: 470-482.
[13]	Gao Chunlei, Yi Weining, Cui Wenyu, et al. Simulation of infrared imaging in different atmospheric conditions [J]. Infrared and Laser Engineering, 2012, 41(3): 258-591. (in Chinese) 高春雷, 易维宁, 崔文煜, 等. 不同大气条件下红外成像效果模拟[J]. 红外与激光工程, 2012, 41(3): 258-591.
[14]
[15]
[16]	Cheng Y. Mean Shift, mode seeking, and clustering [J]. IEEE Trans Pattern Analysis and Machine Intelligence, 1995, 17(8): 790-799.
[17]
[18]	Gao Guowang, Liu Shangqian, Qin Han lin, et al. Robustness tracking algorithm for the infrared target under complex background noise [J]. Journal of Xidian University (Natural Science Edition), 2010, 37(6): 1099-1102. (in Chinese) 高国旺, 刘上乾, 秦翰林, 等. 强背景噪声下红外目标的鲁棒性跟踪算法[J]. 西安电子科技大学学报(自然科学版), 2010, 37(6): 1099-1102.
[19]	Fan Xinnan, Ding Penghua, Liu Junding, et al. Infrared object tracking based on gray and SURF features fusion [J]. Journal of Image and Graphics, 2012, 17(11): 1376-1383. (in Chinese) 范新南, 丁朋华, 刘俊定, 等. 融合灰度和SURF 特征的红外目标跟踪[J]. 中国图形图像学报, 2012, 17(11): 1376-1383.

[1]	杜俊廷, 常冰, 李照宇, 张浩, 秦琛烨, 耿勇, 谭腾, 周恒, 姚佰承. 中红外光学频率梳：进展与应用（特邀） . 红外与激光工程, 2022, 51(3): 20210969-1-20210969-15. doi: 10.3788/IRLA20210969
[2]	王亚平, 周裕丰, 张宝华. 基于去雾增强和张量恢复的红外小目标检测 . 红外与激光工程, 2022, 51(4): 20210417-1-20210417-9. doi: 10.3788/IRLA20210417
[3]	尹丽华, 杭娟, 康亮, 刘士建. 基于联合相机路径的红外视频稳像算法 . 红外与激光工程, 2021, 50(6): 20200405-1-20200405-11. doi: 10.3788/IRLA20200405
[4]	裴晓敏, 范慧杰, 唐延东. 时空特征融合深度学习网络人体行为识别方法 . 红外与激光工程, 2018, 47(2): 203007-0203007(6). doi: 10.3788/IRLA201847.0203007
[5]	尹丽华, 李范鸣, 刘士建. 基于区域分割与融合的全景稳像算法 . 红外与激光工程, 2018, 47(9): 926004-0926004(10). doi: 10.3788/IRLA201847.0926004
[6]	葛宝义, 左宪章, 胡永江, 张岩. 基于双步相关滤波的目标跟踪算法 . 红外与激光工程, 2018, 47(12): 1226004-1226004(10). doi: 10.3788/IRLA201847.1226004
[7]	张冬晓, 何永强, 胡文刚, 陈一超. 自适应红外隐身系统的背景投影建模 . 红外与激光工程, 2016, 45(3): 304001-0304001(7). doi: 10.3788/IRLA201645.0304001
[8]	柯熙政, 邓莉君. 半导体激光器的非线性失真特性及预失真补偿研究 . 红外与激光工程, 2015, 44(11): 3204-3210.
[9]	潘雄, 张春生, 王夏霄, 王熙辰, 赵亚飞, 王定球. 反馈延迟对光纤陀螺振动误差特性的影响 . 红外与激光工程, 2014, 43(8): 2607-2612.
[10]	郭敬明, 何昕, 杨杰, 魏仲慧, 龚俊亮. 模板自适应的Mean Shift红外目标跟踪 . 红外与激光工程, 2014, 43(4): 1087-1093.
[11]	王兆丰, 闫镔, 童莉, 陈健, 李建新. 自适应邻域尺寸选择的点云法向量估计算法 . 红外与激光工程, 2014, 43(4): 1322-1326.
[12]	刘永兴, 张培晴, 戴世勋, 王训四, 林常规, 张巍, 聂秋华, 徐铁峰. 中红外硫系光子晶体光纤参量放大特性模拟研究 . 红外与激光工程, 2014, 43(2): 511-516.
[13]	周强, 赵巨峰, 冯华君, 徐之海, 李奇, 陈跃庭. 基于偏振成像的红外图像增强 . 红外与激光工程, 2014, 43(1): 39-47.
[14]	王华伟, 曹剑中, 马彩文, 张辉, 武登山. 具有自适应校正功能的红外成像系统设计 . 红外与激光工程, 2014, 43(1): 61-66.
[15]	金雁, 江洁, 张广军. 高动态星跟踪方法 . 红外与激光工程, 2013, 42(1): 212-217.
[16]	赵菲, 卢焕章, 张志勇. 滑动窗口核岭回归运动目标轨迹预测算法 . 红外与激光工程, 2013, 42(3): 829-835.
[17]	齐蕾, 张闻文, 陈钱, 顾国华. EMCCD图像自适应模糊中值滤波算法研究 . 红外与激光工程, 2013, 42(11): 3150-3155.
[18]	卢瑞涛, 黄新生, 徐婉莹. 基于Contourlet变换和Facet模型的红外小目标检测方法 . 红外与激光工程, 2013, 42(8): 2281-2287.
[19]	杨咚, 余伟. Sage_Husa自适应滤波在大方位失准角初始对准的研究 . 红外与激光工程, 2013, 42(8): 2197-2201.
[20]	刘运龙, 薛雨丽, 袁素真, 毛峡. 基于局部均值的红外小目标检测算法 . 红外与激光工程, 2013, 42(3): 814-822.

点击查看大图

计量

文章访问数: 371
HTML全文浏览量: 57
PDF下载量: 179
被引次数: 0

基于改进Mean-Shift 算法的红外小目标跟踪

1. 中国科学院西安光学精密机械研究所,陕西西安 710119;

2. 中国科学院大学,北京 100049;

3. 中国科学院国家授时中心,陕西西安 710600

作者简介:
杨一帆（1987-），男，硕士生，主要从事图像目标跟踪方面的研究工作。Email：yangyifan@yeah.net

收稿日期: 2013-11-10
修回日期: 2013-12-13
刊出日期: 2014-07-25

中图分类号: TP391

关键词:

摘要: 复杂背景下的红外小目标跟踪在目标跟踪领域一直是重要的研究方向。由于小目标体量小、机动性大，而红外图像大多受到严重的背景噪声和热噪声影响，使得针对红外小目标的跟踪大多出错率高，鲁棒性不强。针对红外小目标的跟踪，提出了一种改进的Mean-Shift 算法。结合图像的统计特性，提出了一种自适应非线性算法对图像进行处理；同时融合了图像的梯度直方图对目标进行描述。实验通过对高强度噪声和高遮挡环境下视频目标进行跟踪，比较了传统Mean-Shift 算法和改进后算法的跟踪效果，结果显示文中提出的改进算法不但可以有效地跟踪目标，而且大幅降低了跟踪窗口与目标之间的相对抖动，增强了跟踪算法的鲁棒性。

English Abstract

Tracking of infrared small-target based on improved Mean-Shift algorithm

1. Xi'an Institute of Optics and Precision Mechanics Chinese Academy of Sciences,Xi'an 710119,China;

2. University of Chinese Academy of Sciences,Beijing 100049 China;

3. National Time Service Center,Chinese Academy of Sciences,Xi'an,710600,China

Received Date: 2013-11-10
Rev Recd Date: 2013-12-13
Publish Date: 2014-07-25

Keywords:

Abstract: Small-target tracking in infrared imagery with a complex background is always an important task in object tracking fields. Small and manoeuvrable objects in complex clutter and highly noised background usually results in serious false alarm in target tarking for low contrast of infrared imagery. An improved Mean-Shift algorithm to handle the influnce of complex background during tracking the smalltarget in infrared imagery was proposed. This work proposed an adaptive nonlinear machine to help Mean-Shift algorithm to get stable histogram of the interesting areas. This machine expanded the imformation of object histogram refer to the mean value of tracking window, as well as reduced the noise part of tracking window refer to the standard deviation of it. At the same time, algorithm fused the histograms of gradient with histogram of gray-value to discribe the target. To validate the effection of the proposed algorithm, the last part conduct a series tracking experiments which choose highly noised and clutered videos as their candidates. The comparison of the tracking results between tradtional Mean-Shift algotithm and improved Mean-Shift algorithm shows that the proposed algorithm has a more accurate tracking effection. Further more the proposed algorithm highly reduced the wobbleing between small-target and tracking window. This indicates that the improved algorithm achieved more robustness.