A Practical Image Augmentation Method for Construction Safety Using Object Range Expansion Synthesis

This study aims to propose a practical and realistic synthetic data generation method for object recognition in hazardous and data-scarce environments, such as construction sites. Artificial intelligence (AI) applications in such dynamic domains require domain-specific datasets, yet collecting real-...

Full description

Saved in:
Bibliographic Details
Main Authors: Jaemin Kim, Ingook Wang, Jungho Yu, Seulki Lee
Format: Article
Language:English
Published: MDPI AG 2025-04-01
Series:Buildings
Subjects:
Online Access:https://www.mdpi.com/2075-5309/15/9/1447
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850275413316599808
author Jaemin Kim
Ingook Wang
Jungho Yu
Seulki Lee
author_facet Jaemin Kim
Ingook Wang
Jungho Yu
Seulki Lee
author_sort Jaemin Kim
collection DOAJ
description This study aims to propose a practical and realistic synthetic data generation method for object recognition in hazardous and data-scarce environments, such as construction sites. Artificial intelligence (AI) applications in such dynamic domains require domain-specific datasets, yet collecting real-world data can be challenging due to safety concerns, logistical constraints, and high labor costs. To address these limitations, we introduce object range expansion synthesis (ORES), a lightweight and non-generative method for generating synthetic image data by inserting real object masks into varied background scenes using open datasets. ORES synthesizes new scenes, while preserving scale and ground alignment, enabling controllable and realistic data augmentation. A dataset of 30,000 synthetic images was created using the proposed method and used to train an object recognition model. When tested on real-world construction site images, the model achieved a mean average precision at IoU 0.50 (mAP50) of 98.74% and a recall of 54.55%. While recall indicates room for improvement, the high precision highlights the practical value of synthetic data in enhancing model performance without requiring extensive field data collection. This research contributes a scalable approach to data generation in safety-critical and data-deficient environments, reducing dependence on direct data acquisition, while maintaining model efficacy. It provides a foundation for accelerating the deployment of AI technologies in high-risk industries by overcoming data bottlenecks and supporting real-world applications through practical synthetic augmentation.
format Article
id doaj-art-4f8cc9d309df49d1a0bdfb9148ed5369
institution OA Journals
issn 2075-5309
language English
publishDate 2025-04-01
publisher MDPI AG
record_format Article
series Buildings
spelling doaj-art-4f8cc9d309df49d1a0bdfb9148ed53692025-08-20T01:50:45ZengMDPI AGBuildings2075-53092025-04-01159144710.3390/buildings15091447A Practical Image Augmentation Method for Construction Safety Using Object Range Expansion SynthesisJaemin Kim0Ingook Wang1Jungho Yu2Seulki Lee3Department of Architecture Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Architecture Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Architecture Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaDepartment of Architecture Engineering, Kwangwoon University, Seoul 01897, Republic of KoreaThis study aims to propose a practical and realistic synthetic data generation method for object recognition in hazardous and data-scarce environments, such as construction sites. Artificial intelligence (AI) applications in such dynamic domains require domain-specific datasets, yet collecting real-world data can be challenging due to safety concerns, logistical constraints, and high labor costs. To address these limitations, we introduce object range expansion synthesis (ORES), a lightweight and non-generative method for generating synthetic image data by inserting real object masks into varied background scenes using open datasets. ORES synthesizes new scenes, while preserving scale and ground alignment, enabling controllable and realistic data augmentation. A dataset of 30,000 synthetic images was created using the proposed method and used to train an object recognition model. When tested on real-world construction site images, the model achieved a mean average precision at IoU 0.50 (mAP50) of 98.74% and a recall of 54.55%. While recall indicates room for improvement, the high precision highlights the practical value of synthetic data in enhancing model performance without requiring extensive field data collection. This research contributes a scalable approach to data generation in safety-critical and data-deficient environments, reducing dependence on direct data acquisition, while maintaining model efficacy. It provides a foundation for accelerating the deployment of AI technologies in high-risk industries by overcoming data bottlenecks and supporting real-world applications through practical synthetic augmentation.https://www.mdpi.com/2075-5309/15/9/1447synthetic dataartificial intelligence performancemachine learningconstruction sites
spellingShingle Jaemin Kim
Ingook Wang
Jungho Yu
Seulki Lee
A Practical Image Augmentation Method for Construction Safety Using Object Range Expansion Synthesis
Buildings
synthetic data
artificial intelligence performance
machine learning
construction sites
title A Practical Image Augmentation Method for Construction Safety Using Object Range Expansion Synthesis
title_full A Practical Image Augmentation Method for Construction Safety Using Object Range Expansion Synthesis
title_fullStr A Practical Image Augmentation Method for Construction Safety Using Object Range Expansion Synthesis
title_full_unstemmed A Practical Image Augmentation Method for Construction Safety Using Object Range Expansion Synthesis
title_short A Practical Image Augmentation Method for Construction Safety Using Object Range Expansion Synthesis
title_sort practical image augmentation method for construction safety using object range expansion synthesis
topic synthetic data
artificial intelligence performance
machine learning
construction sites
url https://www.mdpi.com/2075-5309/15/9/1447
work_keys_str_mv AT jaeminkim apracticalimageaugmentationmethodforconstructionsafetyusingobjectrangeexpansionsynthesis
AT ingookwang apracticalimageaugmentationmethodforconstructionsafetyusingobjectrangeexpansionsynthesis
AT junghoyu apracticalimageaugmentationmethodforconstructionsafetyusingobjectrangeexpansionsynthesis
AT seulkilee apracticalimageaugmentationmethodforconstructionsafetyusingobjectrangeexpansionsynthesis
AT jaeminkim practicalimageaugmentationmethodforconstructionsafetyusingobjectrangeexpansionsynthesis
AT ingookwang practicalimageaugmentationmethodforconstructionsafetyusingobjectrangeexpansionsynthesis
AT junghoyu practicalimageaugmentationmethodforconstructionsafetyusingobjectrangeexpansionsynthesis
AT seulkilee practicalimageaugmentationmethodforconstructionsafetyusingobjectrangeexpansionsynthesis