Conference proceeding
A generative artificial intelligence benchmarking study: comparing diffusion and GAN-based models for surface crack image generation in civil infrastructure
Proceedings of SPIE, the international society for optical engineering, v 13951, 139510T
16 Apr 2026
Abstract
This study investigates the generation of synthetic surface crack images in civil infrastructure using two state-of-the-art generative artificial intelligence models: the denoising diffusion probabilistic model (DDPM) and StyleGAN3-ada. The objective is to assess their capability to produce realistic crack patterns for potential use in addressing data scarcity in infrastructure monitoring. Both models were trained on two datasets consisting of 9,887 images of surface cracks in concrete and asphalt pavements, including RGB and masked binary versions. The quality of generated images was evaluated using the Fréchet Inception Distance (FID), a metric for measuring the similarity between real and synthetic images. The DDPM achieved lower FID scores of 30.1 for RGB images and 22.4 for masked images, indicating closer alignment with real distributions. In contrast, StyleGAN3-ada obtained higher FID scores of 45.9 for RGB and 52.2 for masked images, reflecting inferior performance. Visual inspection also showed that DDPM-generated images exhibited more realistic textures, orientations, and crack morphologies, whereas StyleGAN3-ada frequently produced blurred or distorted patterns. These findings demonstrate that diffusion-based approaches outperform GAN-based methods in generating surface crack patterns that closely replicate real-world characteristics. The study highlights the effectiveness of DDPM in capturing complex geometrical and textural features of cracks, underscoring its suitability for generating high-fidelity synthetic data.
Metrics
1 Record Views
Details
- Title
- A generative artificial intelligence benchmarking study: comparing diffusion and GAN-based models for surface crack image generation in civil infrastructure
- Creators
- Pedram Bazrafshan - Drexel UniversityArvin Ebrahimkhanlou - Drexel University
- Publication Details
- Proceedings of SPIE, the international society for optical engineering, v 13951, 139510T
- Conference
- SPIE SPIE Smart Structures and Materials + Nondestructive Evaluation (Vancouver, BC, Canada)
- Series
- Proceedings of SPIE
- Publisher
- SPIE
- Number of pages
- 10
- Grant note
- American Society for Nondestructive Testing (ASNT) National Science Foundation (NSF): 2450862 NSF MRI: 2320600
This work was supported in part by the American Society for Nondestructive Testing (ASNT) and National Science Foundation (NSF) Award Number 2450862. Any opinions, findings, and conclusions or recommendations expressed in this paper are those of the authors and do not necessarily reflect the views of sponsors and they have not approved or endorsed its content. The authors also acknowledge Drexel's University Research Computing Facility (URCF) for providing HPC resources that have contributed to the research results reported within this paper. The authors are thankful for the access to the computational resources provided through the NSF MRI Award Number 2320600.
- Resource Type
- Conference proceeding
- Language
- English
- Academic Unit
- Civil, Architectural, and Environmental Engineering; Mechanical Engineering and Mechanics
- Web of Science ID
- WOS:001776710100019
- Scopus ID
- 2-s2.0-105040404530
- Other Identifier
- 991022180806404721