In order to explore the comprehensive discrimination technology of dew and frost observation, a near surface temperature and relative humidity observation system was built in the Atmospheric Detection Test Base of CMA in 2014. Based on the nearsurface temperature and humidity observations, automatic station data, manual recordings, and video recordings of dew and frost, the comprehensive discrimination technology of dew and frost is investigated. The difference of grass temperature and nearsurface dewpoint temperature is preliminarily selected as the main distinguishing factor of dew and frost. Dew and frost occur when the difference of grass temperature and nearsurface dewpoint temperature is less than a threshold, which is related to the lowest temperature of the day. Data analysis shows that the discrimination results of dew and frost by using the comprehensive indexes are close to the actual observations, with the concordance rate being 84.5%. The nearsurface temperature and relative humidity have good indication of dew and frost.