Cargando…

Influence of Insufficient Dataset Augmentation on IoU and Detection Threshold in CNN Training for Object Detection on Aerial Images

The objects and events detection tasks are being performed progressively often by robotic systems like unmanned aerial vehicles (UAV) or unmanned surface vehicles (USV). Autonomous operations and intelligent sensing are becoming standard in numerous scenarios such as supervision or even search and r...

Descripción completa

Detalles Bibliográficos
Autores principales: Bożko, Arkadiusz, Ambroziak, Leszek
Formato: Online Artículo Texto
Lenguaje:English
Publicado: MDPI 2022
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9740240/
https://www.ncbi.nlm.nih.gov/pubmed/36501781
http://dx.doi.org/10.3390/s22239080
_version_ 1784848011941117952
author Bożko, Arkadiusz
Ambroziak, Leszek
author_facet Bożko, Arkadiusz
Ambroziak, Leszek
author_sort Bożko, Arkadiusz
collection PubMed
description The objects and events detection tasks are being performed progressively often by robotic systems like unmanned aerial vehicles (UAV) or unmanned surface vehicles (USV). Autonomous operations and intelligent sensing are becoming standard in numerous scenarios such as supervision or even search and rescue (SAR) missions. The low cost of autonomous vehicles, vision sensors and portable computers allows the incorporation of the deep learning, mainly convolutional neural networks (CNN) in these solutions. Many systems meant for custom purposes rely on insufficient training datasets, what may cause a decrease of effectiveness. Moreover, the system’s accuracy is usually dependent on the returned bounding boxes highlighting the supposed targets. In desktop applications, precise localisation might not be particularly relevant; however, in real situations, with low visibility and non-optimal camera orientation, it becomes crucial. One of the solutions for dataset enhancement is its augmentation. The presented work is an attempt to evaluate the influence of the training images augmentation on the detection parameters important for the effectiveness of neural networks in the context of object detection. In this research, network appraisal relies on the detection confidence and bounding box prediction accuracy (IoU). All the applied image modifications were simple pattern and colour alterations. The obtained results imply that there is a measurable impact of the augmentation process on the localisation accuracy. It was concluded that a positive or negative influence is related to the complexity and variability of the objects classes.
format Online
Article
Text
id pubmed-9740240
institution National Center for Biotechnology Information
language English
publishDate 2022
publisher MDPI
record_format MEDLINE/PubMed
spelling pubmed-97402402022-12-11 Influence of Insufficient Dataset Augmentation on IoU and Detection Threshold in CNN Training for Object Detection on Aerial Images Bożko, Arkadiusz Ambroziak, Leszek Sensors (Basel) Article The objects and events detection tasks are being performed progressively often by robotic systems like unmanned aerial vehicles (UAV) or unmanned surface vehicles (USV). Autonomous operations and intelligent sensing are becoming standard in numerous scenarios such as supervision or even search and rescue (SAR) missions. The low cost of autonomous vehicles, vision sensors and portable computers allows the incorporation of the deep learning, mainly convolutional neural networks (CNN) in these solutions. Many systems meant for custom purposes rely on insufficient training datasets, what may cause a decrease of effectiveness. Moreover, the system’s accuracy is usually dependent on the returned bounding boxes highlighting the supposed targets. In desktop applications, precise localisation might not be particularly relevant; however, in real situations, with low visibility and non-optimal camera orientation, it becomes crucial. One of the solutions for dataset enhancement is its augmentation. The presented work is an attempt to evaluate the influence of the training images augmentation on the detection parameters important for the effectiveness of neural networks in the context of object detection. In this research, network appraisal relies on the detection confidence and bounding box prediction accuracy (IoU). All the applied image modifications were simple pattern and colour alterations. The obtained results imply that there is a measurable impact of the augmentation process on the localisation accuracy. It was concluded that a positive or negative influence is related to the complexity and variability of the objects classes. MDPI 2022-11-23 /pmc/articles/PMC9740240/ /pubmed/36501781 http://dx.doi.org/10.3390/s22239080 Text en © 2022 by the authors. https://creativecommons.org/licenses/by/4.0/Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
spellingShingle Article
Bożko, Arkadiusz
Ambroziak, Leszek
Influence of Insufficient Dataset Augmentation on IoU and Detection Threshold in CNN Training for Object Detection on Aerial Images
title Influence of Insufficient Dataset Augmentation on IoU and Detection Threshold in CNN Training for Object Detection on Aerial Images
title_full Influence of Insufficient Dataset Augmentation on IoU and Detection Threshold in CNN Training for Object Detection on Aerial Images
title_fullStr Influence of Insufficient Dataset Augmentation on IoU and Detection Threshold in CNN Training for Object Detection on Aerial Images
title_full_unstemmed Influence of Insufficient Dataset Augmentation on IoU and Detection Threshold in CNN Training for Object Detection on Aerial Images
title_short Influence of Insufficient Dataset Augmentation on IoU and Detection Threshold in CNN Training for Object Detection on Aerial Images
title_sort influence of insufficient dataset augmentation on iou and detection threshold in cnn training for object detection on aerial images
topic Article
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9740240/
https://www.ncbi.nlm.nih.gov/pubmed/36501781
http://dx.doi.org/10.3390/s22239080
work_keys_str_mv AT bozkoarkadiusz influenceofinsufficientdatasetaugmentationoniouanddetectionthresholdincnntrainingforobjectdetectiononaerialimages
AT ambroziakleszek influenceofinsufficientdatasetaugmentationoniouanddetectionthresholdincnntrainingforobjectdetectiononaerialimages