Based on the differences in material properties between the target and the background, and combined with changes in environmental, lighting, and temporal conditions, an optical imaging feature prediction model for ship targets in a port background is constructed to explore the impact of the "thermal crossover" phenomenon on the infrared detection performance of ship targets. Firstly, Fluent software is used to calculate the surface temperature distribution of the target under different environmental conditions. Combined with the material and radiation scattering characteristics of different parts of the target, a radiation transmission model of the ship target is constructed. Secondly, Landsat 8 remote sensing data is used to invert the port temperature distribution under specific seasonal and temporal conditions, and land cover classification is performed based on a supervised classification algorithm. Based on meteorological data, the Kriging algorithm is used to achieve temperature expansion of the port background under different temporal conditions. Meanwhile, topographic factors such as elevation, slope, and aspect of land features are taken into account to establish a high-precision temporal temperature reconstruction model with topographic modulation. The differences in sea surface radiation characteristics and scattering effects are taken into account to establish a background radiation transfer model. Finally, combining solar, sky, and atmospheric radiative transfer models, the modulation effect of the space-based optical platform is considered, and an imaging prediction model of the "ship target-port background-ambient illumination-atmospheric transmission" is established. The results show that during non-thermal crossover periods in summer, the detectability of 3-5μm is better than that of 8-14μm, while the opposite is true during thermal crossover periods in winter. At the critical moment of thermal crossover in summer, the detectability of 3-5μm is worse than that of 8-14μm, while at the critical moment of thermal crossover in winter, the detectability of 3-5μm and 8-14μm are relatively close.
YUAN et al. (Mon,) studied this question.