Inverse thermography via Monte Carlo infrared rendering and simultaneous sensitivities estimation
Overview & Synopsis
Infrared imaging is widely used to retrieve temperatures and material properties, but its interpretation is strongly affected by multi-reflection effects and uncertainties in surface emissivity, especially in complex environments such as tokamak devices and urban areas. In this work, we address these challenges by reformulating the infrared rendering equation into a fully probabilistic, recursive form and by introducing a Monte Carlo estimator capable of simultaneously computing both pixel-wise intensity and its sensitivities with respect to any non-geometric parameter. Our formulation preserves the full geometric complexity of the scene and follows an analogue, pathspace interpretation that enables physical insight and straightforward multi-physics coupling. We demonstrate that the same sampled paths can be used to estimate intensities and sensitivities without additional computational cost. The approach is validated on benchmark configurations and applied to two inverse problems: temperature retrieval with fixed surface properties (a linear case) and Bidirectional Reflectance Distribution Function (BRDF) parameter retrieval with fixed temperatures (a nonlinear case). In both examples, the sensitivities enable classical inversion methods (least squares and Levenberg-Marquardt) to recover the parameters with good accuracy. The results highlight the potential of Monte Carlo infrared rendering with simultaneous sensitivities estimation as a foundation for robust and physically consistent inverse thermography.
hal.science)
← / → for pages • + / - for zoom • 0 to fit width
Powered by Mozilla PDF.js v3.11
Document Information
| Document Title | Inverse thermography via Monte Carlo infrared rendering and simultaneous sensitivities estimation |
|---|---|
| Author / Organization | Zili He, Fabrice Rigollet, Anna Glasser, Eric Serre, Jonathan Gaspar, Léa Penazzi |
| Subject & Discipline | Multidisciplinary Academic Open Research |
| Categories | Science & Mathematics |
| Publication Year | 2026 |
| License / Rights | Open Access / Creative Commons |
| Resource Verification | Listed PDF (Verification Pending) |
| Original Host Website |
HAL Open Science
(hal.science)
|
| File Format | application/pdf (Portable Document Format) |
| Language | EN |
| Catalog ID | DOC-123883 |
| Original Source Link | Visit Source Page on hal.science » |