Conference proceeding
An Overview of Deep Learning in UAV Perception
2024 IEEE International Conference on Consumer Electronics (ICCE), pp 1-6
06 Jan 2024
Abstract
Detecting objects autonomously from a drone camera presents significant challenges. One major obstacle is the lack of an adequate dataset in terms of both quantity and diversity. Additionally, current algorithms have primarily focused on finding objects at close range, which limits their efficacy in effectively avoiding collisions with obstacles at longer distances. Our goal here is to conduct a comprehensive survey of various state-of the-art drone-captured imagery datasets and drone detection algorithms, evaluating their adequacy for enabling computer vision in safe autonomous drone operations. In this aim, we analyzed sixteen drone based image datasets and twenty-four one object detection algorithms. For a concise summary table of the reviewed methodologies and databases, please refer to the link provided: https://research.coe.drexel.edu/ece/imaple/380-2/
Metrics
12 Record Views
3 citations in Scopus
Details
- Title
- An Overview of Deep Learning in UAV Perception
- Creators
- Amirreza Rouhi - Drexel UniversitySolmaz Arezoomandan - Drexel UniversityRitik Kapoor - Drexel UniversityJohn Klohoker - Drexel UniversitySneh Patal - Drexel UniversityPrincie Shah - West Windsor-Plainsboro High School South,West Windsor,NJ,USAHimanshu Umare - Drexel UniversityDavid Han - Drexel University
- Publication Details
- 2024 IEEE International Conference on Consumer Electronics (ICCE), pp 1-6
- Publisher
- IEEE
- Number of pages
- 6
- Grant note
- Federal Aviation Administration (10.13039/100006282)
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Electrical and Computer Engineering
- Scopus ID
- 2-s2.0-85186973399
- Other Identifier
- 991021931086404721