Compensation for the effects of limited field of view coverage on infrared thermometer measurements

Infrared thermometers play a pivotal role in temperature measurement across various industrial fields due to their convenient utilization and cost-efficient solution. However, the accuracy of measurements remains constrained, especially as the influence of environmental radiation becomes pronounced with restricted field-of-view (FOV) coverage. To address such challenges, this work demonstrates a model-based compensation strategy to mitigate the environmental radiation impact encountered in scenarios with low FOV coverage. For more complex cases, the model is complemented with a data-driven strategy to enhance its adaptivity and computational efficiency. Experimental validation demonstrates the model’s effectiveness, reducing the error margin from an initial level exceeding 15% to below 4% across various conditions, including measuring distance, central offset, actual temperature, and angular inclination. In human hand testing, the hybrid compensation method achieves superior temperature measurement accuracy (0.11 ± 0.05 °C), outperforming conventional nonlinear regression and deep learning approaches. The developed compensation approach, with its demonstrated effectiveness and versatility, opens up a myriad of opportunities to improve the accuracy and reliability of low-cost infrared thermometer usage. • Modeling for limited field-of-view effects enhances the accuracy of infrared thermometry. • The model cuts measurement errors down to 4% from 15% in varied conditions. • Integrating data-driven methods addresses complex scenarios effectively.