Volume 38, Number 6, December 2020
|Page(s)||1146 - 1153|
|Published online||02 February 2021|
Aerial Target Tracking Based on Frequency-Domain Gabor Filters and Correlation Filters under Complex Interference Environment
School of Astronautics, Northwestern Polytechnical University, Xi'an 710072, China
Aerial infrared target automatic tracking technology is the core technology of the photoelectric countermeasure system of infrared imaging missiles. Aiming at challenging conditions of complex battlefield environment, large scale change of air target, infrared interference partial occlusion and so on, we propose a new correlation filtering algorithm based on Gabor filtering features in the frequency domain for aerial infrared target tracking. This algorithm constructs a set of frequency-domain Gabor filters, which extracts frequency-domain Gabor filtering(GF) features of the target image block and reduces dimension fusion, thereby effectively suppressing background noise and highlighting target texture information. According to the characteristics and variation law of spectrum energy distribution, the eigenvectors of target spectrum scale are extracted to improve the accuracy of target scale information estimation. High-confidence patches are used to track the reliable part of the target, and the energy distribution of the tracking box is calculated to predict the occluded part of the target simultaneously. This method provides more target information and improves the robustness of the algorithm in case of reappearance of the occluded target. Compared with other tracking algorithms, the average accuracy of the proposed algorithm was improved by 15.2%, and the frame frequency reached above 110 Hz.
Key words: aerial infrared target / frequency-domain Gabor filter / correlation filter / high-confidence patches
关键字 : 空中红外目标 / 频域Gabor滤波 / 相关滤波 / 高置信分块
© 2020 Journal of Northwestern Polytechnical University. All rights reserved.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.