Effects of Driving Background Complexity and Interface Opacity on Visual Cognition in AR-HUD Systems

The cognitive effectiveness of AR-HUD interfaces is influenced by driving background complexity (DBC) and information opacity. This study explores how they impact visual cognition and reaction efficiency using a dual-phase experimental approach. In Experiment I, a subjective evaluation classified DBC into low, medium, and high levels based on static driving scene images. This was followed by an objective assessment of the complexity of color variety, edge density, and texture features for the selected L-DBC, M-DBC, and H-DBC images. Experiment II then employed eye-tracking metrics (reaction time, mean pupil diameter, and AOI fixation duration) to evaluate participants' visual performance across 10 opacity gradients (0.1–1.0). Results revealed significant interactions between DBC and opacity levels. Under L-DBC, M-DBC, and H-DBC conditions, the relationship between information opacity and reaction times exhibited different phases. To optimize visual cognitive performance, AR-HUD opacity should be set at a minimum of 0.6. When opacity levels are below 0.7, the greater the DBC, the longer the response time for the same opacity. When the information opacity is above 0.7, quicker reaction times can be achieved, regardless of whether the DBC is high or low. These findings offer valuable design guidelines for optimizing AR-HUD text opacity in complex driving backgrounds.