Euclidean consistency-driven dual-layer information fusion framework for UAV-based traffic accident scene reconstruction
Zhihao Xie, Wenjing Xia, Cheng He, Taifeng Qiu · 2026
Read Euclidean consistency-driven dual-layer information fusion framework for UAV-based traffic accident scene reconstruction by Zhihao Xie, Wenjing Xia, Cheng He, Taifeng Qiu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0350987.txt.
Description
Zhihao Xie and co-authors tuned how drones should fly to rebuild a traffic accident scene in three dimensions. Investigators photograph crash sites from the air, but altitude and photo overlap decide whether the model is true to centimeters or warped. Using a two-layer analysis that pairs a plain statistical model with a learning model, they tested twenty-seven flight setups against measured errors.
Altitude and overlap worked together rather than separately, and the two analysis layers agreed strongly, which supports the approach. A sweet spot emerged: flying low, roughly twenty to twenty-five meters up, with heavy forward overlap and slightly less side overlap gave centimeter-level accuracy with stable geometry. A trial at a real crash scene backed the recipe. For police and insurers, the settings turn drone photos into evidence-grade models.
Catalog record
| CREATOR | Zhihao Xie, Wenjing Xia, Cheng He, Taifeng Qiu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 48545 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |