Lightweight Detection Network Based on Sub-Pixel Convolution and Objectness-Aware Structure for UAV Images
Unmanned Aerial Vehicles (UAVs) can serve as an ideal mobile platform in various situations. Real-time object detection with on-board apparatus provides drones with increased flexibility as well as a higher intelligence level. In order to achieve good detection results in UAV images with complex gro...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2021-08-01
|
Series: | Sensors |
Subjects: | |
Online Access: | https://www.mdpi.com/1424-8220/21/16/5656 |
_version_ | 1797521980661956608 |
---|---|
author | Xuanye Li Hongguang Li Yalong Jiang Meng Wang |
author_facet | Xuanye Li Hongguang Li Yalong Jiang Meng Wang |
author_sort | Xuanye Li |
collection | DOAJ |
description | Unmanned Aerial Vehicles (UAVs) can serve as an ideal mobile platform in various situations. Real-time object detection with on-board apparatus provides drones with increased flexibility as well as a higher intelligence level. In order to achieve good detection results in UAV images with complex ground scenes, small object size and high object density, most of the previous work introduced models with higher computational burdens, making deployment on mobile platforms more difficult.This paper puts forward a lightweight object detection framework. Besides being anchor-free, the framework is based on a lightweight backbone and a simultaneous up-sampling and detection module to form a more efficient detection architecture. Meanwhile, we add an objectness branch to assist the multi-class center point prediction, which notably improves the detection accuracy and only takes up very little computing resources. The results of the experiment indicate that the computational cost of this paper is 92.78% lower than the CenterNet with ResNet18 backbone, and the mAP is 2.8 points higher on the Visdrone-2018-VID dataset. A frame rate of about 220 FPS is achieved. Additionally, we perform ablation experiments to check on the validity of each part, and the method we propose is compared with other representative lightweight object detection methods on UAV image datasets. |
first_indexed | 2024-03-10T08:23:04Z |
format | Article |
id | doaj.art-099f9960851c4e8cbf056304aa51382a |
institution | Directory Open Access Journal |
issn | 1424-8220 |
language | English |
last_indexed | 2024-03-10T08:23:04Z |
publishDate | 2021-08-01 |
publisher | MDPI AG |
record_format | Article |
series | Sensors |
spelling | doaj.art-099f9960851c4e8cbf056304aa51382a2023-11-22T09:43:35ZengMDPI AGSensors1424-82202021-08-012116565610.3390/s21165656Lightweight Detection Network Based on Sub-Pixel Convolution and Objectness-Aware Structure for UAV ImagesXuanye Li0Hongguang Li1Yalong Jiang2Meng Wang3School of Electrical and Information Engineering, Beihang University, Beijing 100191, ChinaUnmanned System Research Institute, Beihang University, Beijing 100191, ChinaUnmanned System Research Institute, Beihang University, Beijing 100191, ChinaSchool of Electrical and Information Engineering, Beihang University, Beijing 100191, ChinaUnmanned Aerial Vehicles (UAVs) can serve as an ideal mobile platform in various situations. Real-time object detection with on-board apparatus provides drones with increased flexibility as well as a higher intelligence level. In order to achieve good detection results in UAV images with complex ground scenes, small object size and high object density, most of the previous work introduced models with higher computational burdens, making deployment on mobile platforms more difficult.This paper puts forward a lightweight object detection framework. Besides being anchor-free, the framework is based on a lightweight backbone and a simultaneous up-sampling and detection module to form a more efficient detection architecture. Meanwhile, we add an objectness branch to assist the multi-class center point prediction, which notably improves the detection accuracy and only takes up very little computing resources. The results of the experiment indicate that the computational cost of this paper is 92.78% lower than the CenterNet with ResNet18 backbone, and the mAP is 2.8 points higher on the Visdrone-2018-VID dataset. A frame rate of about 220 FPS is achieved. Additionally, we perform ablation experiments to check on the validity of each part, and the method we propose is compared with other representative lightweight object detection methods on UAV image datasets.https://www.mdpi.com/1424-8220/21/16/5656lightweight convolutional neural networkobject detectionUAV images |
spellingShingle | Xuanye Li Hongguang Li Yalong Jiang Meng Wang Lightweight Detection Network Based on Sub-Pixel Convolution and Objectness-Aware Structure for UAV Images Sensors lightweight convolutional neural network object detection UAV images |
title | Lightweight Detection Network Based on Sub-Pixel Convolution and Objectness-Aware Structure for UAV Images |
title_full | Lightweight Detection Network Based on Sub-Pixel Convolution and Objectness-Aware Structure for UAV Images |
title_fullStr | Lightweight Detection Network Based on Sub-Pixel Convolution and Objectness-Aware Structure for UAV Images |
title_full_unstemmed | Lightweight Detection Network Based on Sub-Pixel Convolution and Objectness-Aware Structure for UAV Images |
title_short | Lightweight Detection Network Based on Sub-Pixel Convolution and Objectness-Aware Structure for UAV Images |
title_sort | lightweight detection network based on sub pixel convolution and objectness aware structure for uav images |
topic | lightweight convolutional neural network object detection UAV images |
url | https://www.mdpi.com/1424-8220/21/16/5656 |
work_keys_str_mv | AT xuanyeli lightweightdetectionnetworkbasedonsubpixelconvolutionandobjectnessawarestructureforuavimages AT hongguangli lightweightdetectionnetworkbasedonsubpixelconvolutionandobjectnessawarestructureforuavimages AT yalongjiang lightweightdetectionnetworkbasedonsubpixelconvolutionandobjectnessawarestructureforuavimages AT mengwang lightweightdetectionnetworkbasedonsubpixelconvolutionandobjectnessawarestructureforuavimages |