To improve the efficiency and accuracy of atmospheric transmittance parameter acquisition in infrared radiation testing, a model for atmospheric transmittance calculation based on the HO-RF algorithm is proposed. Using key environmental factors such as temperature, humidity, operating distance, and atmospheric pressure as inputs, a regression model is established based on the measured data, enabling rapid and accurate atmospheric transmittance calculation. The simulation results show that the proposed model outperforms traditional back propagation (BP) neural network and random forest (RF) models in terms of accuracy, with a root mean square error (RMSE) reduced to 0.010745, an R² value of 0.95877, and a mean absolute error (MAE) of 0.0080021. This model effectively reduces experimental complexity and outperforms traditional methods in terms of accuracy, stability, and reliability. It can improve the efficiency of infrared characteristic testing for fighter aircraft and has excellent practical application value.