Vehicle Detection Method in UAV Aerial Video

Authors

Suqin Wang, 1. School of Control and Computer Engineering, North China Electric Power University, Beijing 102206, China;;
Wenhao Shi, 1. School of Control and Computer Engineering, North China Electric Power University, Beijing 102206, China;;2. Institute of Computing Technology, Beijing 100190, China;
Zhaoxin Li, 2. Institute of Computing Technology, Beijing 100190, China;
Tianlu Mao, 2. Institute of Computing Technology, Beijing 100190, China;

Abstract

Abstract: The use of UAV aerial photography to obtain traffic information has many advantages like a flexible high mobility, comprehensive information and so on, which plays a vital role in the emergency response, anti-terrorism/battlefield and other environments. This paper proposes a method for vehicle detection in UAV aerial video. Aiming at the problem of UAV video image deflection, the image registration method based on SURF algorithm is used to correct the deflection of UAV hatching video according to the standard frame. The background and the moving vehicle are separated by the pixel-based adaptive segmentation algorithm to detect the moving vehicle in the video. Experiments show that the matching accuracy of our method rate up to 90%, and can detect the movement of vehicles accurately in a large range of scenes, whose recall(Re), precision(Pr), false negative rate (FNR) are 82.3%, 90.1%, 17.3%.

Recommended Citation

Wang, Suqin; Shi, Wenhao; Li, Zhaoxin; and Mao, Tianlu (2019) "Vehicle Detection Method in UAV Aerial Video," Journal of System Simulation: Vol. 30: Iss. 7, Article 43.
DOI: 10.16182/j.issn1004731x.joss.201807043
Available at: https://dc-china-simulation.researchcommons.org/journal/vol30/iss7/43

First Page

2776

DOI Link

https://doi.org/10.16182/j.issn1004731x.joss.201807043

Last Page

2786

CLC

TP391

Recommended Citation

Wang Suqin, Shi Wenhao, Li Zhaoxin, Mao Tianlu. Vehicle Detection Method in UAV Aerial Video[J]. Journal of System Simulation, 2018, 30(7): 2776-2786.

DOI

10.16182/j.issn1004731x.joss.201807043