Abstract: The reference standard system of blackbody visibility is composed of industrial camera, blackbody and industrial personal computer. With the industrial camera lens facing south, shooting blackbody and sky background, the chosen house is 680 meters away from the camera and the window of the house faces north. A series of extinction processing are finished to ensure the house as a blackbody. The camera takes pictures of the blackbody every 6 seconds. The mathematical model of the reference standard system of blackbody is established according to the definition of meteorological visibility and the Cosimi De's law, which shows the relationship between the meteorological visibility and sky/blackbody luminosity, and the relationship between the meteorological visibility and the blackbody blackness or the industrial camera image uniformity at the same time. Using the mathematical model, the error caused by the blackbody blackness and image uniformity of the camera is analyzed. When visibility is less than 30 km, the error caused by blackbody blackness and CCD industrial camera's uniformity of the measured system is about 3.7%. Blackbody blackness is measured by photographic method, putting a standard white card under the blackbody window, and the camera lens is 600 mm away from the window, shooting the white card and window at the same time. The brightness of the window is taken as the light-emission luminance, and the brightness white card divided by 0.7332 is taken as the incident light intensity. The ratio between them is the brightness, and the blackness of measurements for the house is 0.0018. Integrating sphere is used for uniformity and linear calibration of the camera. Compared with the forward scattering visibility instrument measurement and the reference standard system of blackbody visibility, the value of the visibility is consistent, but is greater under high visibility condition, and the value of the blackbody visibility measurement is lower under low visibility condition. Perhaps the cause is that the principle of the forward scattering visibility instrument cannot measure atmospheric aerosol absorption. Under low visibility condition, the aerosol optical absorption effect is larger, and the forward scattering visibility instrument cannot measure the absorption attenuation and measurements may be higher. To confirm this conclusion, further solubility and synchronous observation of aerosol optical properties are needed.