Object Recognition Through UAV Observations Based on Yolo and Generative Adversarial Network

Bo Li, Zhigang Gan, Evgeny Sergeevich Neretin, Zhipeng Yang

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

3 Scopus citations

Abstract

Aiming at the object recognition through UAV, an intelligent object recognition model based on YOLO and Generative adversarial network is proposed in this paper. Firstly, the solution is given, and an object recognition model that can realize intelligent recognition is established. Then, in order to improve the resolution of the identified images, an image resolution enhancement model based on generative adversarial networks is built. After that, the structure and parameters of the recognition model and image resolution enhancement model are adjusted through the simulation experiments to improve the accuracy and robustness of the object recognition. Finally, the object recognition model based on YOLO and generative adversarial network in this paper is verified through UAV.

Original languageEnglish
Title of host publicationIoT as a Service - 6th EAI International Conference, IoTaaS 2020, Proceedings
EditorsBo Li, Changle Li, Mao Yang, Zhongjiang Yan, Jie Zheng
PublisherSpringer Science and Business Media Deutschland GmbH
Pages439-449
Number of pages11
ISBN (Print)9783030675134
DOIs
StatePublished - 2021
Event6th EAI International Conference on IoT as a Service, IoTaaS 2020 - Xi'an, China
Duration: 19 Nov 202020 Nov 2020

Publication series

NameLecture Notes of the Institute for Computer Sciences, Social-Informatics and Telecommunications Engineering, LNICST
Volume346
ISSN (Print)1867-8211
ISSN (Electronic)1867-822X

Conference

Conference6th EAI International Conference on IoT as a Service, IoTaaS 2020
Country/TerritoryChina
CityXi'an
Period19/11/2020/11/20

Keywords

  • Machine learning
  • Object recognition
  • UAV

Fingerprint

Dive into the research topics of 'Object Recognition Through UAV Observations Based on Yolo and Generative Adversarial Network'. Together they form a unique fingerprint.

Cite this