Development of a virtual environment for rapid generation of synthetic training images for artificial intelligence object recognition

Simple item page
dc.contributor.authorWang, Chenyu
dc.contributor.authorTinsley, Lawrence
dc.contributor.authorHonarvar Shakibaei Asli, Barmak
dc.date.accessioned2025-01-09T16:07:02Z
dc.date.available2025-01-09T16:07:02Z
dc.date.freetoread2025-01-09
dc.date.issued2024-12-01
dc.date.pubOnline2024-11-29
dc.description.abstractIn the field of machine learning and computer vision, the lack of annotated datasets is a major challenge for model development and accuracy improvement. Synthetic data generation addresses this issue by providing large, diverse, and accurately annotated datasets, thereby enhancing model training and validation. This study presents a Unity-based virtual environment that utilises the Unity Perception package to generate high-quality datasets. First, high-precision 3D (Three-Dimensional) models are created using a 3D structured light scanner, with textures processed to remove specular reflections. These models are then imported into Unity to generate diverse and accurately annotated synthetic datasets. The experimental results indicate that object recognition models trained with synthetic data achieve a high rate of performance on real images, validating the effectiveness of synthetic data in improving model generalisation and application performance. Monocular distance measurement verification shows that the synthetic data closely matches real-world physical scales, confirming its visual realism and physical accuracy.
dc.description.journalNameElectronics
dc.identifier.citationWang C, Tinsley L, Honarvar Shakibaei Asli B. (2024) Development of a virtual environment for rapid generation of synthetic training images for artificial intelligence object recognition. Electronics, Volume 13, Issue 23, December 2024, Article number 4740
dc.identifier.eissn2079-9292
dc.identifier.elementsID560117
dc.identifier.issn1450-5843
dc.identifier.issueNo23
dc.identifier.paperNo4740
dc.identifier.urihttps://doi.org/10.3390/electronics13234740
dc.identifier.urihttps://dspace.lib.cranfield.ac.uk/handle/1826/23340
dc.identifier.volumeNo13
dc.languageEnglish
dc.language.isoen
dc.publisherMDPI
dc.publisher.urihttps://www.mdpi.com/2079-9292/13/23/4740
dc.rightsAttribution 4.0 Internationalen
dc.rights.urihttp://creativecommons.org/licenses/by/4.0/
dc.subject40 Engineering
dc.subject4009 Electronics, Sensors and Digital Hardware
dc.subjectBioengineering
dc.subjectMachine Learning and Artificial Intelligence
dc.subjectNetworking and Information Technology R&D (NITRD)
dc.subject4009 Electronics, sensors and digital hardware
dc.titleDevelopment of a virtual environment for rapid generation of synthetic training images for artificial intelligence object recognition
dc.typeArticle
dc.type.subtypeJournal Article
dcterms.dateAccepted2024-11-27

Files

Original bundle
Now showing 1 - 1 of 1
Thumbnail Image
Name:
Development_of_a_virtual_environment-2024.pdf
Size:
119.19 MB
Format:
Adobe Portable Document Format
Download
License bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
license.txt
Size:
1.63 KB
Format:
Plain Text
Description:
Download

Collections

Staff publications (SATM)