Research on real time tracking algorithm of multispectral mosaic image based on the fast DSST algorithm

Yanming Shao; Dapeng Li; Xiaotong Zhang; Qianjin Lin; Huabao Long

doi:10.1117/12.2580259

5 November 2020 Research on real time tracking algorithm of multispectral mosaic image based on the fast DSST algorithm

Yanming Shao, Dapeng Li, Xiaotong Zhang, Qianjin Lin, Huabao Long

Proceedings Volume 11567, AOPC 2020: Optical Sensing and Imaging Technology; 115673W (2020) https://doi.org/10.1117/12.2580259
Event: Applied Optics and Photonics China (AOPC 2020), 2020, Beijing, China

Abstract

In order to solve the problem of traditional target recognition and tracking algorithms of the multispectral image such as high computation complexity, poor real time performance and low stability under complex scene and great variation of target appearance, a new mosaic image tracking algorithm based on dimension reduction of HOG feature data and multi-scale correlation filter is proposed in this paper. Firstly, in order to reduce the calculation complexity as well as to enhance the detection rate of small target, the 2D multispectral mosaic image data instead of the traditional 3D multispectral image data cubes is used, Then the histogram of oriented gradient (HOG) feature is extracted from the mosaic image data, and the singular value decomposition (SVD) algorithm with improved threshold selection method is adopted to reduce the dimension of the HOG feature matrix. Compared to the method which extracts HOG feature after dimension reduction, the proposed method takes advantage of high recognition precision, simple operation and high real-time performance. Finally, the target tracking is realized based on the dimension-reduced HOG feature with the fast discriminative scale space tracker (fDSST) algorithm which combines the scale filter and the position filter. A multispectral image dataset for target tracking was established, including different target occlusion, motion blur, variation of target scale and target appearance. Target tracking results on the dataset show the proposed algorithm can realize good tracking continuity and stability even if there exist different ground objects, variation in the appearance of the target shape, or target reappearance after occlusion.

Citation Download Citation

Yanming Shao, Dapeng Li, Xiaotong Zhang, Qianjin Lin, and Huabao Long "Research on real time tracking algorithm of multispectral mosaic image based on the fast DSST algorithm", Proc. SPIE 11567, AOPC 2020: Optical Sensing and Imaging Technology, 115673W (5 November 2020); https://doi.org/10.1117/12.2580259

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
7 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Detection and tracking algorithms

Multispectral imaging

Target detection

Image filtering

Dimension reduction

Target recognition

Cameras

Show All Keywords

Keywords/Phrases

Search In:

Publication Years