Research on target loss early warning of KCF algorithm based on hypothesis test

Ying Zhang; Chuan-gang Xu; Feng Zhang; Li-quan Song

doi:10.1117/12.2504958

12 December 2018 Research on target loss early warning of KCF algorithm based on hypothesis test

Ying Zhang, Chuan-gang Xu, Feng Zhang, Li-quan Song

Proceedings Volume 10846, Optical Sensing and Imaging Technologies and Applications; 108461K (2018) https://doi.org/10.1117/12.2504958
Event: International Symposium on Optoelectronic Technology and Application 2018, 2018, Beijing, China

Abstract

The KCF algorithm uses the response values of the template and the test sample to calculate the target position. The location of the maximum response value is the target region. However, the KCF algorithm does not set the target loss warning mechanism. The target will be lost when the tracking process encounters complicated conditions such as scale chance, rapid movement, and severe obstruction. At this point, the tracker will update the background information into the template. The accumulated deviation of the template will cause the tracker cannot be correctly identified and tracked when it encounters the target again. To solve this problem, through the statistics of the maximum response value of each frame of the video sequence, it founds that the overall response of the video sequence has a normal distribution trend, and the maximum response value of the frame will fluctuate abnormally in the frames that are lost or are about to be lost. This paper uses the idea of hypothesis testing in mathematical statistics. This paper uses a set of fixed-dimensional response peak data to perform hypothesis testing on the peak value of the current frame response. If the response peak of the current frame falls within the rejection region, the target is determined to be lost or is about to be lost. The experimental results prove that the proposed method can correctly implement the early warning function when verifying the OTB standard test sequence. It can provide reference for when to load redetection after target is lost.

Citation Download Citation

Ying Zhang, Chuan-gang Xu, Feng Zhang, and Li-quan Song "Research on target loss early warning of KCF algorithm based on hypothesis test", Proc. SPIE 10846, Optical Sensing and Imaging Technologies and Applications, 108461K (12 December 2018); https://doi.org/10.1117/12.2504958

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
6 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

Target detection

Video

Optical tracking

Image filtering

Algorithm development

Chemical elements

Show All Keywords

Keywords/Phrases

Search In:

Publication Years