An infrared image generation method based on physical effects modeling

Infrared images play a crucial role in performance evaluation of infrared systems and the development of image processing algorithms. However, acquiring high-quality infrared images frequently encounters substantial difficulties stemming from both hardware constraints and environmental factors. To overcome these limitations, we present a physically-based computational imaging framework for generating synthetic infrared imagery. This methodology rigorously simulates the end-to-end infrared imaging chain, integrating two fundamental physical processes: atmospheric radiative transfer, and infrared system effects. Through rigorous integration of radiative transfer physics, spatial modulation effects, and noise mechanisms, we establish a physically accurate infrared imaging model. Experimental results demonstrate the proposed model achieves excellent agreement with empirical measurement data, accurately capturing the essential imaging characteristics of infrared systems. This work provides critical support for infrared camera design optimization and performance evaluation.