What question did this study set out to answer?

Develop an automated three-phase framework using YOLOv12 and machine learning to recognize pipeline categories and quantify burial depth and diameter from ground-penetrating radar (GPR) data.

August 19, 2026Journal of Computing in Civil Engineering

Automatic Recognition of Category, Burial Depth, and Diameter for Underground Pipeline Using GPR Based on YOLOv12 and Machine Learning Algorithm

View Full Paper

Ask AI

Bookmark

Authors

FMFengbo Ma JGJiachen Gao HPHaifeng Pang

Discussion

Member takes

Overview

Algorithmic study demonstrates automated detection of underground pipeline type, depth, and diameter from radar data, highlighting improved efficiency for urban subsurface infrastructure mapping.

Key Points

Develop an automated three-phase framework using YOLOv12 and machine learning to recognize pipeline categories and quantify burial depth and diameter from ground-penetrating radar (GPR) data.
Implemented a sequential pipeline combining YOLOv12 for object detection in GPR B-scan images, hyperbolic curve feature extraction, and CatBoost regression modeling.
Trained and evaluated the system across four utility classes: plastic-water, electric-cable, plastic-gas, and steel-water.
YOLOv12 achieved a mean average precision (mAP@0.5:0.95) of 0.699, with category APs of 0.649 (plastic–water), 0.654 (electric-cable), 0.755 (plastic–gas), and 0.737 (steel–water).
Hyperbolic feature extraction showed standard deviations of relative errors of 2.9% for vertex vertical coordinate and 12.4% for vertex curvature.
CatBoost regression predicted burial depth and diameter with R² scores of 0.9973 and 0.9425, MAE values of 0.0154 m and 0.0185 m, and MAPE values of 1.38% and 8.41%, respectively.

Cite This Study

Ma et al. (2026) studied this question.

synapsesocial.com/papers/6a85636403308d306e2d685b https://doi.org/10.1061/jccee5.cpeng-7853

View Full Paper

Ask AI

Bookmark

August 19, 2026Journal of Computing in Civil Engineering

Automatic Recognition of Category, Burial Depth, and Diameter for Underground Pipeline Using GPR Based on YOLOv12 and Machine Learning Algorithm

View Full Paper

Ask AI

Bookmark

Authors

FMFengbo Ma JGJiachen Gao HPHaifeng Pang

Discussion

Member takes

Overview

Algorithmic study demonstrates automated detection of underground pipeline type, depth, and diameter from radar data, highlighting improved efficiency for urban subsurface infrastructure mapping.

Key Points

Develop an automated three-phase framework using YOLOv12 and machine learning to recognize pipeline categories and quantify burial depth and diameter from ground-penetrating radar (GPR) data.
Implemented a sequential pipeline combining YOLOv12 for object detection in GPR B-scan images, hyperbolic curve feature extraction, and CatBoost regression modeling.
Trained and evaluated the system across four utility classes: plastic-water, electric-cable, plastic-gas, and steel-water.
YOLOv12 achieved a mean average precision (mAP@0.5:0.95) of 0.699, with category APs of 0.649 (plastic–water), 0.654 (electric-cable), 0.755 (plastic–gas), and 0.737 (steel–water).
Hyperbolic feature extraction showed standard deviations of relative errors of 2.9% for vertex vertical coordinate and 12.4% for vertex curvature.
CatBoost regression predicted burial depth and diameter with R² scores of 0.9973 and 0.9425, MAE values of 0.0154 m and 0.0185 m, and MAPE values of 1.38% and 8.41%, respectively.

Cite This Study

Ma et al. (2026) studied this question.

synapsesocial.com/papers/6a85636403308d306e2d685b https://doi.org/10.1061/jccee5.cpeng-7853

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark