Fast Automatic Vehicle Detection in UAV Images Using Convolutional Neural Networks

Xin Luo; Xiaoyue Tian; Huijie Zhang; Weimin Hou; Geng Leng; Wenbo Xu; Haitao Jia; Xixu He; Meng Wang; Jian Zhang

doi:10.3390/rs12121994

Remote Sensing (Jun 2020)

Fast Automatic Vehicle Detection in UAV Images Using Convolutional Neural Networks

Xin Luo,
Xiaoyue Tian,
Huijie Zhang,
Weimin Hou,
Geng Leng,
Wenbo Xu,
Haitao Jia,
Xixu He,
Meng Wang,
Jian Zhang

Affiliations

Xin Luo: Spatial Information and Digital Technology, School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
Xiaoyue Tian: Spatial Information and Digital Technology, School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
Huijie Zhang: Department of Communication Engineering, School of Automation and Electrical Engineering, Chengdu Technological University, Chengdu 611730, China
Weimin Hou: Department of Communication Engineering, School of Information Science and Engineering, Hebei University of Science and Technology, Shijiazhuang 050018, China
Geng Leng: Spatial Information and Digital Technology, School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
Wenbo Xu: Spatial Information and Digital Technology, School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
Haitao Jia: Spatial Information and Digital Technology, School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
Xixu He: Spatial Information and Digital Technology, School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
Meng Wang: Spatial Information and Digital Technology, School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China
Jian Zhang: Spatial Information and Digital Technology, School of Resources and Environment, University of Electronic Science and Technology of China, Chengdu 611731, China

DOI: https://doi.org/10.3390/rs12121994
Journal volume & issue: Vol. 12, no. 12
p. 1994

Abstract

Read online

Vehicle targets in unmanned aerial vehicle (UAV) images are generally small, so a significant amount of detailed information on targets may be lost after neural computing, which leads to the poor performances of the existing recognition algorithms. Based on convolutional neural networks that utilize the YOLOv3 algorithm, this article focuses on the development of a quick automatic vehicle detection method for UAV images. First, a vehicle dataset for target recognition is constructed. Then, a novel YOLOv3 vehicle detection framework is proposed according to the following characteristics: The vehicle targets in the UAV image are relatively small and dense. The average precision (AP) increased by 5.48%, from 92.01% to 97.49%, which still remains the rather high processing speed of the YOLO network. Finally, the proposed framework is tested using three datasets: COWC, VEDAI, and CAR. The experimental results demonstrate that our method had a better detection capability.

Published in Remote Sensing

ISSN: 2072-4292 (Online)
Publisher: MDPI AG
Country of publisher: Switzerland
LCC subjects: Science
Website: http://www.mdpi.com/journal/remotesensing/

About the journal

Abstract

Keywords