||The study aims to explore into the visibility of text information on large LED display as the effective segment information, quick search and correct reading performance. The experiment sample design is conducted on three variables, namely color combination of figure/background, tracking and leading. The experiment is conducted in three stages, including “Accuracy Recognition Measure,” “Distance Subjective Evaluation and Glare Level Questionnaire” and “Glare Distance Measure.” The data analysis adopts descriptive statistics analysis and multi-factor analysis of variance. The results suggest that color combination of figure/background, tracking and leading all have impact on glare distance but do not affect each other. The furthest glare distance is red/green, which is the combination that is more likely to perceive glare, followed by red/red. The closest distance is yellow/blue, which is the combination that is less likely to perceive glare. Findings of objective recognition and subjective evaluation indicate that the color combination with high recognition accuracy is not necessarily identical to visual subjective perception. In sum of the results between subjective and objective comparison, the best combination of visibility for color combination of figure/background is yellow/blue, amber/amber, and yellow/yellow. With regards to tracking, the standard glare distance of tracking is greater than tracking fine-tuned to 100 and is likely to perceive glare. With regards to leading, leading 150% is better than leading 120%. Tracking and leading show significant difference at 28 meter, the visibility of leading 150% is better than that of leading 120% and tracking fine-tuned to 100 is better than the tracking standard. Nonetheless, tracking at 44 meters does not show significance. The study concludes the design principles that meet the public visual perception, which results can be referred for the information on LED display design in extensive applications such as the train station, shops, airport, or roads.