AI-Powered Physics-Informed Correlation Technologies for Advanced Thermal-Imaging in Near and Far Fields

We introduce a generalized Dual-Probe imaging system integrating in a unified platform both Infra-Red (IR) and Thermoreflectance (TR) lenses featuring coated-patterning. A dedicated, diffusion-aware, calibration-kit (CK) with a scalable array of resizable grid, seen as collectively interacting patterned unit-cells under random-field fluctuations, is introduced. Powered by an adaptive biasing DC supply, the CK module includes standard connectivity for easy interfacing through a simple command set, making it compatible with standard programming environments (e.g., Matlab, C, Python, Labview, and others). With PID control technology, it guarantees real-time temperature compensation as a function of airflow. The module also offers a user-friendly graphical interface to ensure an effortless setup process for facilitating a parametric AI-powered database creation and indexation for thermal-visual image identification and cross-correlations targeting advanced multi-physics characterization and industrial-testing applications. The proposed multi-probing system, referenced as “Cognitive-Lens”, integrates correlation-based signal processing powered by physics-informed AI algorithms for creating real-time instruments and tools applicable to noise and variation-aware on-the-move imaging applications. Perspectives are drawn for Multi-Scale and Super-Resolution technologies with sub-micron spatial resolution (beyond diffraction limit).