Robust Infrared Target Tracking Based on Histograms of Sparse Coding

被引：0
|
作者：
Yang F. [1 ]
Yang D. [1 ]
Mao N. [1 ]
Li X. [1 ]
机构：
[1] School of Control Science and Engineering, Hebei University of Technology, Tianjin
来源：
Guangxue Xuebao | / 11卷
关键词：
Distractor-aware model; Histogram of sparse coding; Infrared image; Machine vision; Target tracking;
D O I：
10.3788/AOS201737.1115002
中图分类号：
学科分类号：
摘要：
Making use of information in infrared images to build an effective observation model is the basis for realizing robust infrared target tracking. Besides the regular factors that have adverse influence on visual target tracking, infrared target tracking is faced with other difficulties as well, such as lack of edge and texture information, low signal-to-noise ratio and background clutter. An infrared target tracking algorithm based on histograms of sparse coding (HSC) and the distractor-aware model (DAM) is proposed, which exploits K singular value decomposition algorithm to obtain an overcomplete dictionary. With the dictionary, sparse code of every pixel is computed to compose HSC as a descriptor, and DAM is utilized to strengthen resistance against background clutter. The proposed algorithm does not only use structural information of tracked object but also eliminates the influence of background clutter. Compared with other tracking algorithms, the proposed algorithm achieves 3.8% and 4.4% enhancement on VOT-TIR2015 dataset with respect to precision and success rate, respectively, possessing high research and practical value. © 2017, Chinese Lasers Press. All right reserved.
引用
收藏
相关论文
共 27 条
  • [1] He Y.J., Li M., Zhang J.L., Et al., Infrared target tracking via weighted correlation filter, Infrared Physics & Technology, 73, pp. 103-114, (2015)
  • [2] Huang Y., Zheng H., Yang H., Improving an object tracker for infrared flying bird tracking, International Conference on Image Processing, pp. 1699-1703, (2016)
  • [3] Demir S., Cetin E., Co-difference based object tracking algorithm for infrared videos, International Conference on Image Processing, pp. 434-438, (2016)
  • [4] Zhang H., Zhao B., Tang L., Et al., Infrared object tracking based on adaptive multi-features integrations, Acta Optica Sinica, 30, 5, pp. 1291-1296, (2010)
  • [5] Guan Z., Chen Q., Qian W., Et al., Infrared target tracking algorithm based on algorithm fusion, Acta Optica Sinica, 28, 5, pp. 860-865, (2008)
  • [6] Sun J., Ding Y., Zhou L., Visually interactive hand tracking algorithm combined with infrared depth tracking, Acta Optica Sinica, 37, 1, (2017)
  • [7] Zhao A., Wang H., Yang X., Et al., An affine invariant method of forward looking infra-red target recognition, Laser & Optoelectronics Progress, 52, 7, (2015)
  • [8] Bao C.L., Wu Y., Ling H.B., Et al., Real time robust l1 tracker using accelerated proximal gradient approach, IEEE Conference on Computer Vision and Pattern Recognition, pp. 1830-1837, (2012)
  • [9] Danelljan M., Khan F.S., Felsberg M., Et al., Adaptive color attributes for real-time visual tracking, IEEE Conference on Computer Vision and Pattern Recognition, pp. 1090-1097, (2014)
  • [10] Zhang K.H., Zhang L., Liu Q.S., Et al., Fast visual tracking via dense spatio-temporal context learning, European Conference on Computer Vision, pp. 127-141, (2014)
123