ThermoGridX — High-Res Thermal Field Reconstruction

Overview

A MATLAB pipeline that reconstructs a high-resolution thermal map from coarse, sparse sensor readings — turning 5cm-spaced measurements into a continuous 0.5cm-resolution surface, a 10× improvement in spatial detail useful anywhere sensor density is physically limited.

Architecture

Coarse gridded sensor input → 2D interpolation (interp2) → high-resolution continuous thermal surface reconstruction across a 20×20cm target area.

Key Features

• 10× spatial resolution improvement via interp2-based reconstruction.
• Continuous temperature distribution rendering across the full 20×20cm surface.
• Direct applicability to thermal imaging and engineering diagnostics with sparse sensor placement.

Development Process

• Simulated coarse 5cm-spaced sensor data as the interpolation input.
• Applied interp2 to reconstruct a continuous high-resolution surface.
• Validated the reconstructed surface against expected thermal gradients.

Challenges & Solutions

Results & Impact

The reconstructed thermal map achieved a 10× spatial accuracy improvement over the raw sensor grid, demonstrating practical value for imaging and diagnostic applications with limited sensor density.

Future Improvements

• Compare interp2 against higher-order interpolation methods.
• Apply to real sensor hardware data.
• Add uncertainty visualization for interpolated regions.