Face gears are a promising new type of transmission in the field of crossed-axis drive systems. Considering the elastohydrodynamic lubrication of rough surfaces is crucial for enhancing their performance. Therefore, this paper proposes a simulation method for line contact hydrodynamic lubrication that accurately characterizes the actual processed surface morphology. First, we compared two methods for characterizing gear surface morphology: fractal theory and FFT filtering. The results demonstrate that fractal theory has significant advantages in accurately representing gear surface morphology at different scales. Additionally, we extracted surface parameters from white light interferometry data using the structure function method, achieving precise modeling of gear surface roughness. Based on this, we established a line contact hydrodynamic lubrication model that accounts for roughness, allowing for a quantitative analysis of changes in oil film thickness and pressure distribution under different roughness conditions. Finally, case studies showed that an increase in gear surface roughness leads to greater fluctuations in oil film thickness and pressure, with a decreasing trend in minimum film thickness. However, in certain localized areas, the complexity of the rough surface can cause an increase in film thickness. This finding provides valuable guidance for the design and optimization of face gear transmission systems, helping to enhance lubrication performance through effective control of surface morphology.