Visibility data have long needed to traffic meteorological monitoring and warning system, but visibility data have monitored with expensive special equipment. Visibility degradation in fog is due to the light scattering of fog droplets, which are transit from aerosols via activation. Considering strong correlation between PM2.5 (Particulate matter with diameters less than 2.5μm) mass concentration and visibility, regression models can be useful tools for retrieving visibility data from available PM2.5 data. In this study, PM2.5 is measured by low cost and commercial equipment. The results of experiment indicate that relative humidity is the key factor to impact accuracy correlation between PM2.5 and visibility, the strongest correlation locates in the RH (<60%). Results of the studies suggest that visibility decreases with increases of PM2.5 mass concentration; however, it has been found the decrease rate tapers off gradually. In order to capture the real-time visibility data, to grasp the process of low visibility events, the design of remote monitoring system is put forward. Using the GPRS network to link to cloud as a server, proposed the Arduino as the controller, design and implements a wireless serial acquisition and control system based LabVIEW and Arduino, this system can achieve the function of real-time synchronization Web publishing. The result of the test indicates that this system has typical characteristics of friendly interface, high levels of reliability and expansibility, moreover it can retrieve visibility data from available PM2.5 data that can easy to access by low-cost sensor along the highway.