Speaker
Description
Dynamic thermal infrared imaging remains a challenging task for uncooled microbolometer-based cameras, primarily due to their limited thermal time constants and the resulting motion-induced blur. In order to investigate these effects, we present a compact dynamic laboratory test bench designed to emulate the passage of hot objects from the camera perspective without physically moving heated masses.
The setup relies on stationary heated panels and a self-designed rotating disk featuring apertures. A dedicated design study was conducted to define the angular width and geometry of the disk apertures, taking into account the rotational speed constraints of the brushless motor and the frame rate of the infrared cameras. This design ensures that motion-induced blur remains confined within the projected zones of the thermal panels, allowing its reliable quantification.
The thermal sources consist of two independently heated panels built from high thermal conductivity material to ensure rapid and uniform heat distribution. The panels are covered by a small layer of Nextel Velvet 811-21 paint to ensure a stable and known temperature in the heating range used. Heating is achieved using MINCO HAP6948-1 polyimide heating elements, each integrating a PT1000 temperature sensor. Temperature regulation is performed through a closed-loop PID control scheme implemented using a MAX31865PMB1 RTD-to-digital converter, enabling repeatable temperature control from ambient conditions up to 50 °C. Both the heating elements and the motor are controlled through a Wi-Fi-enabled interface, allowing real-time monitoring and parameter adjustment via a smartphone.
The proposed test bench is used to evaluate and compare two uncooled infrared microbolometer cameras operating at the same spatial resolution. One camera corresponds to a conventional uncooled microbolometer, while the second is a new prototype, built in the BRIGHTER’s framework project, referred to as Fast-Pixel, featuring a thermal time constant reduced by a factor of two, equal to 5 ms. A quantitative image analysis is performed to evaluate motion-induced blur and contrast attenuation under dynamic conditions. Several image quality metrics, including SSIM, RMS contrast, and Laplacian variance, are extracted by comparison with static reference images to characterize spatial image degradation and to assess the impact of the detector thermal time constant on dynamic imaging performance.
Acknowledgments :
BRIGHTER project has received funding from the Chips Joint Undertaking (Chips JU) under grant agreement N°101096985. The JU receives support from the European Union’s Horizon Europe research and innovation program and France, Belgium, Portugal, Spain, Turkey
