Layered difference-based background removal for infrared thermal imagery of aero-engine nozzle and exhaust flow

In the testing of infrared characteristics in aero-engine exhaust flow experiments, the application of infrared thermal imaging is crucial, and the removal of background interference in infrared thermal images is a key step to ensure data accuracy. This paper proposes a background removal method for aero-engine nozzle and exhaust flow infrared thermal image based on hierarchical differencing, aiming to enhance the efficiency and precision of infrared characteristic test data processing. The method first converts the measured infrared thermal image into a relative energy map, then calculates energy curves and energy difference curves through hierarchical processing. By inversely traversing the energy difference curve, the background cutoff layer is effectively identified. Finally, a connected region algorithm is employed to remove discrete interference points, resulting in a clean target image. Based on the radiation variation characteristics of infrared images, the proposed method achieves effective separation of background and target, offering advantages of stability, speed, and automation, along with adjustable parameters and broad applicability. This technology not only significantly improves the quality of infrared characteristic test data processing but also provides robust technical support for optimizing and evaluating the infrared stealth performance of aero-engines.