A synthetic segmentation dataset generator using a 3D modeling framework and raycaster: a mining industry application

Many industries utilize deep learning methods to increase efficiency and reduce costs. One of these methods, image segmentation, is used for object detection and recognition in localization and mapping. Segmentation models are trained using labeled datasets; however, manually creating datasets for e...

Full description

Saved in:
Bibliographic Details
Main Authors: Wilhelm Johannes Kilian, Jaco Prinsloo, Jan Vosloo, Stéphan Taljaard
Format: Article
Language:English
Published: Frontiers Media S.A. 2024-12-01
Series:Frontiers in Artificial Intelligence
Subjects:
Online Access:https://www.frontiersin.org/articles/10.3389/frai.2024.1453931/full
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850061456753557504
author Wilhelm Johannes Kilian
Jaco Prinsloo
Jan Vosloo
Stéphan Taljaard
author_facet Wilhelm Johannes Kilian
Jaco Prinsloo
Jan Vosloo
Stéphan Taljaard
author_sort Wilhelm Johannes Kilian
collection DOAJ
description Many industries utilize deep learning methods to increase efficiency and reduce costs. One of these methods, image segmentation, is used for object detection and recognition in localization and mapping. Segmentation models are trained using labeled datasets; however, manually creating datasets for every application, including deep-level mining, is time-consuming and typically expensive. Recently, many papers have shown that using synthetic datasets (digital recreations of real-world scenes) for training produces highly-accurate segmentation models. This paper proposes a synthetic segmentation dataset generator using a 3D modeling framework and raycaster. The generator was applied to a deep-level mining case study and produced a dataset containing labeled images of scenes typically found in this environment, therefore removing the requirement to create the dataset manually. Validation showed high accuracy segmentation after model training using the generated dataset (compared to other applications that use real-world datasets). Furthermore, the generator can be customized to produce datasets for many other applications.
format Article
id doaj-art-7dcf8ccb1ea448108c8790904d3a7e41
institution DOAJ
issn 2624-8212
language English
publishDate 2024-12-01
publisher Frontiers Media S.A.
record_format Article
series Frontiers in Artificial Intelligence
spelling doaj-art-7dcf8ccb1ea448108c8790904d3a7e412025-08-20T02:50:14ZengFrontiers Media S.A.Frontiers in Artificial Intelligence2624-82122024-12-01710.3389/frai.2024.14539311453931A synthetic segmentation dataset generator using a 3D modeling framework and raycaster: a mining industry applicationWilhelm Johannes KilianJaco PrinslooJan VoslooStéphan TaljaardMany industries utilize deep learning methods to increase efficiency and reduce costs. One of these methods, image segmentation, is used for object detection and recognition in localization and mapping. Segmentation models are trained using labeled datasets; however, manually creating datasets for every application, including deep-level mining, is time-consuming and typically expensive. Recently, many papers have shown that using synthetic datasets (digital recreations of real-world scenes) for training produces highly-accurate segmentation models. This paper proposes a synthetic segmentation dataset generator using a 3D modeling framework and raycaster. The generator was applied to a deep-level mining case study and produced a dataset containing labeled images of scenes typically found in this environment, therefore removing the requirement to create the dataset manually. Validation showed high accuracy segmentation after model training using the generated dataset (compared to other applications that use real-world datasets). Furthermore, the generator can be customized to produce datasets for many other applications.https://www.frontiersin.org/articles/10.3389/frai.2024.1453931/fulldeep learningcomputer visionimage segmentationdeep-level miningreal applications in engineering
spellingShingle Wilhelm Johannes Kilian
Jaco Prinsloo
Jan Vosloo
Stéphan Taljaard
A synthetic segmentation dataset generator using a 3D modeling framework and raycaster: a mining industry application
Frontiers in Artificial Intelligence
deep learning
computer vision
image segmentation
deep-level mining
real applications in engineering
title A synthetic segmentation dataset generator using a 3D modeling framework and raycaster: a mining industry application
title_full A synthetic segmentation dataset generator using a 3D modeling framework and raycaster: a mining industry application
title_fullStr A synthetic segmentation dataset generator using a 3D modeling framework and raycaster: a mining industry application
title_full_unstemmed A synthetic segmentation dataset generator using a 3D modeling framework and raycaster: a mining industry application
title_short A synthetic segmentation dataset generator using a 3D modeling framework and raycaster: a mining industry application
title_sort synthetic segmentation dataset generator using a 3d modeling framework and raycaster a mining industry application
topic deep learning
computer vision
image segmentation
deep-level mining
real applications in engineering
url https://www.frontiersin.org/articles/10.3389/frai.2024.1453931/full
work_keys_str_mv AT wilhelmjohanneskilian asyntheticsegmentationdatasetgeneratorusinga3dmodelingframeworkandraycasteraminingindustryapplication
AT jacoprinsloo asyntheticsegmentationdatasetgeneratorusinga3dmodelingframeworkandraycasteraminingindustryapplication
AT janvosloo asyntheticsegmentationdatasetgeneratorusinga3dmodelingframeworkandraycasteraminingindustryapplication
AT stephantaljaard asyntheticsegmentationdatasetgeneratorusinga3dmodelingframeworkandraycasteraminingindustryapplication
AT wilhelmjohanneskilian syntheticsegmentationdatasetgeneratorusinga3dmodelingframeworkandraycasteraminingindustryapplication
AT jacoprinsloo syntheticsegmentationdatasetgeneratorusinga3dmodelingframeworkandraycasteraminingindustryapplication
AT janvosloo syntheticsegmentationdatasetgeneratorusinga3dmodelingframeworkandraycasteraminingindustryapplication
AT stephantaljaard syntheticsegmentationdatasetgeneratorusinga3dmodelingframeworkandraycasteraminingindustryapplication