Related Experiment Video
Updated: Jun 27, 2026
11:34
Subsurface Defect Localization by Structured Heating Using Laser Projected Photothermal Thermography
Published on: May 15, 2017
3D Localization of Heat Sources Using LiDAR-Thermal Data Fusion and Multisensor Calibration.
Rafał Gasz1, Mateusz Pluskota2, Krzysztof Schwierz2
1Department of Artificial Intelligence, Faculty of Computer Science, Opole University of Technology, Prószkowska 76 St., 45-758 Opole, Poland.
Sensors (Basel, Switzerland)
|June 26, 2026
Summary
This study presents a practical framework for fusing LiDAR and thermal sensing to accurately locate heat sources in 3D. The system achieves precise spatial localization of thermal anomalies without complex machine learning, enhancing robotics and infrastructure monitoring.
Area of Science:
- Robotics and Sensor Fusion
- Computer Vision and Perception
Background:
- LiDAR provides 3D geometry, thermal cameras provide temperature.
- Integrating both enables enhanced 3D scene reconstruction and heat source localization.
- Existing methods may be computationally intensive or require extensive training data.
Purpose of the Study:
- To develop a practical framework for LiDAR-thermal fusion for 3D heat source localization.
- To enable accurate spatial localization of temperature anomalies.
- To provide a robust solution for applications with limited computational resources or training data.
Main Methods:
- Developed a framework integrating Ouster OS1 LiDAR and FLIR A70 thermal camera.
- Implemented intrinsic/extrinsic calibration, data synchronization, and point projection.
- Introduced a novel calibration target for low-contrast LWIR imaging.
Main Results:
- Achieved reprojection errors below 1 pixel.
- Demonstrated a mean hottest-point localization error of ~4.1 cm at 12.3 m distance.
- Validated the framework's accuracy and practicality under laboratory conditions.
Conclusions:
- Accurate spatial localization of thermal anomalies is achievable via geometry-based multimodal fusion.
- The proposed framework is practical, deterministic, and suitable for resource-constrained environments.
- The system has potential applications in building diagnostics, industrial inspection, and robotic perception.
