Low visibility is one of the inducing factors for expressway accidents. Traditional visibility detection devices are usually erected in a complex manner, and are expensive and difficult to be popularized. In order to real-time monitoring the low visibility of whole expressway, a new detection method based on the low visibility visual area proportion of existing monitor image was proposed. Usually, the visual scene area of image reduces with the decrease of its visibility. Therefore, according to the geometric projection relation, visibility could be estimated according to the visual area proportion of image. Mean⁃ while infrared image was used as the reference of common image, which could simplify the processing algorithm, reduce the quality requirement of the common image and improve the practicality of the detection method. Simulation experiment showed that the visual area proportion detection method of infrared image can improve the detection precision, for we only consider the low visibility in the range of 200m, which can greatly affect the driving safety of expressway, and the average error is less than 10%. Therefore, it can be used for high speed real-time monitoring of low visibility of expressway to improve traffic safety.