Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3D traffic accident reconstruction
Abstract The daily occurrence of traffic accidents has led to the development of 3D reconstruction as a key tool for reconstruction, investigation, and insurance claims. This study proposes a novel virtual-real-fusion simulation framework that integrates traffic accident generation, unmanned aerial...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Springer
2024-12-01
|
| Series: | Complex & Intelligent Systems |
| Subjects: | |
| Online Access: | https://doi.org/10.1007/s40747-024-01693-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832571198291050496 |
|---|---|
| author | Yanzhan Chen Qian Zhang Fan Yu |
| author_facet | Yanzhan Chen Qian Zhang Fan Yu |
| author_sort | Yanzhan Chen |
| collection | DOAJ |
| description | Abstract The daily occurrence of traffic accidents has led to the development of 3D reconstruction as a key tool for reconstruction, investigation, and insurance claims. This study proposes a novel virtual-real-fusion simulation framework that integrates traffic accident generation, unmanned aerial vehicle (UAV)-based image collection, and a 3D traffic accident reconstruction pipeline with advanced computer vision techniques and unsupervised 3D point cloud clustering algorithms. Specifically, a micro-traffic simulator and an autonomous driving simulator are co-simulated to generate high-fidelity traffic accidents. Subsequently, a deep learning-based reconstruction method, i.e., 3D Gaussian splatting (3D-GS), is utilized to construct 3D digitized traffic accident scenes from UAV-based image datasets collected in the traffic simulation environment. While visual rendering by 3D-GS struggles under adverse conditions like nighttime or rain, a clustering parameter stochastic optimization model and mixed-integer programming Bayesian optimization (MIPBO) algorithm are proposed to enhance the segmentation of large-scale 3D point clouds. In the numerical experiments, 3D-GS produces high-quality, seamless, and real-time rendered traffic accident scenes achieve a structural similarity index measure of up to 0.90 across different towns. Furthermore, the proposed MIPDBO algorithm exhibits a remarkably fast convergence rate, requiring only 3–5 iterations to identify well-performing parameters and achieve a high $${R}^{2}$$ R 2 value of 0.8 on a benchmark cluster problem. Finally, the Gaussian Mixture Model assisted by MIPBO accurately separates various traffic elements in the accident scenes, demonstrating higher effectiveness compared to other classical clustering algorithms. |
| format | Article |
| id | doaj-art-ce12aa1dc1c94886ad9cb34b16ebc5b0 |
| institution | Kabale University |
| issn | 2199-4536 2198-6053 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | Springer |
| record_format | Article |
| series | Complex & Intelligent Systems |
| spelling | doaj-art-ce12aa1dc1c94886ad9cb34b16ebc5b02025-02-02T12:49:24ZengSpringerComplex & Intelligent Systems2199-45362198-60532024-12-0111112310.1007/s40747-024-01693-9Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3D traffic accident reconstructionYanzhan Chen0Qian Zhang1Fan Yu2Central South UniversityJiangsu Open UniversityCentral South UniversityAbstract The daily occurrence of traffic accidents has led to the development of 3D reconstruction as a key tool for reconstruction, investigation, and insurance claims. This study proposes a novel virtual-real-fusion simulation framework that integrates traffic accident generation, unmanned aerial vehicle (UAV)-based image collection, and a 3D traffic accident reconstruction pipeline with advanced computer vision techniques and unsupervised 3D point cloud clustering algorithms. Specifically, a micro-traffic simulator and an autonomous driving simulator are co-simulated to generate high-fidelity traffic accidents. Subsequently, a deep learning-based reconstruction method, i.e., 3D Gaussian splatting (3D-GS), is utilized to construct 3D digitized traffic accident scenes from UAV-based image datasets collected in the traffic simulation environment. While visual rendering by 3D-GS struggles under adverse conditions like nighttime or rain, a clustering parameter stochastic optimization model and mixed-integer programming Bayesian optimization (MIPBO) algorithm are proposed to enhance the segmentation of large-scale 3D point clouds. In the numerical experiments, 3D-GS produces high-quality, seamless, and real-time rendered traffic accident scenes achieve a structural similarity index measure of up to 0.90 across different towns. Furthermore, the proposed MIPDBO algorithm exhibits a remarkably fast convergence rate, requiring only 3–5 iterations to identify well-performing parameters and achieve a high $${R}^{2}$$ R 2 value of 0.8 on a benchmark cluster problem. Finally, the Gaussian Mixture Model assisted by MIPBO accurately separates various traffic elements in the accident scenes, demonstrating higher effectiveness compared to other classical clustering algorithms.https://doi.org/10.1007/s40747-024-01693-9Traffic accident reconstruction3D Gaussian splattingPoint cloud clusteringExpensive optimizationDeep Gaussian process |
| spellingShingle | Yanzhan Chen Qian Zhang Fan Yu Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3D traffic accident reconstruction Complex & Intelligent Systems Traffic accident reconstruction 3D Gaussian splatting Point cloud clustering Expensive optimization Deep Gaussian process |
| title | Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3D traffic accident reconstruction |
| title_full | Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3D traffic accident reconstruction |
| title_fullStr | Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3D traffic accident reconstruction |
| title_full_unstemmed | Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3D traffic accident reconstruction |
| title_short | Transforming traffic accident investigations: a virtual-real-fusion framework for intelligent 3D traffic accident reconstruction |
| title_sort | transforming traffic accident investigations a virtual real fusion framework for intelligent 3d traffic accident reconstruction |
| topic | Traffic accident reconstruction 3D Gaussian splatting Point cloud clustering Expensive optimization Deep Gaussian process |
| url | https://doi.org/10.1007/s40747-024-01693-9 |
| work_keys_str_mv | AT yanzhanchen transformingtrafficaccidentinvestigationsavirtualrealfusionframeworkforintelligent3dtrafficaccidentreconstruction AT qianzhang transformingtrafficaccidentinvestigationsavirtualrealfusionframeworkforintelligent3dtrafficaccidentreconstruction AT fanyu transformingtrafficaccidentinvestigationsavirtualrealfusionframeworkforintelligent3dtrafficaccidentreconstruction |