Urban Fog-haze Detection Algorithm Based On The Technology Of Data Fusion Research

In In recent years, Fog-haze is increasing, which has seriously jeopardized people’s normal life. So far, most of researches about fog-haze are based on qualitative analysis, which focus on its impact. The research on detection algorithm or the causes of fog-haze is relatively few. The quantitative results on effect of non-meteorological factors for fog-haze is not clear. The definition of fog-haze is not clear. To solve the above problem, the paper puts forward a new theoretical detection algorithm of urban fog-haze, which analyzes problems quantitatively.This article obtains real-time air quality data access from a web page. This paper is based on extensive researched data. The relationship between air particulates and meteorological factors as well as non-meteorological factors was gotten through static correlation and dynamic correlation. Meteorological factors include wind speed, temperature, humidity and pressure. The non-meteorological factors include sulfur dioxide and nitrogen dioxide. By means of PM10 and PM2.5, air quality index(AQI) correlation analysis, it is concluded that PM10 and PM2.5, PM10 and air quality index is positive correlation, so the paper regards PM10 as evaluation index of fog-haze. The forming process of fog-haze is regarded as a black system in this paper. So the paper can ignore the mechanism of forming process of fog-haze. The forming process of fog-haze is just considered from input and output of the system. Then the theory of urban fog-haze detection algorithm is gotten combined with discriminant analysis theory. And the detection algorithm obtained corrects the problem of unrealistically measure in high humidity.In addition, the urban fog-haze levels are defined in this paper based on PM10. The FCM algorithm is applied to the urban fog-haze detection for the first time. The FCM algorithm and MED algorithm are combined with meteorological information, non-meteorological information and other indicators proposed. Thus a new detection algorithm appears.The results of experiments have batter showed that the two detection algorithms of fog-haze proposed in this paper. Moreover, the ability of algorithms is given by comparatively analyzing. The paper gives the scope of the two detection algorithms the detection algorithms.In a word, the two detection algorithms provide a theoretical basis for future more in-depth study of urban haze detection.