Research on construction site dust detection based on prior knowledge MinMax k-Means

Abstract

With the rapid development of urbanization, promoting the process of industrialization has become the best breakthrough to accelerate economic development. The number of construction projects under construction is increasing gradually. In recent years, people are paying more and more attention to the impact of atmospheric particulate matter on the environment and human health. Construction dust is a by-product of open-air construction activities, which does great harm to the ecological environment and human health. It contributes 13.1 % [1]to urban PM2.5 pollution and is also one of the main sources of atmospheric pollutant PM10[2]. In order to timely detect construction site dust and improve the ability of government supervision departments to monitor construction dust pollution, a construction site dust detection method based on prior knowledge Minmax K-means clustering algorithm was proposed. In the process of clustering, the weight which is proportional to the variance in the cluster can be automatically corrected, and the priori knowledge is introduced to deal with the problem that the clustering results are sensitive to the initial position of the clustering center. In addition, the preprocessing adopts the method that the mean value of image blocks with dust is larger than that without dust, and scans the mean value matrix from vertical and horizontal directions to judge whether the image blocks have dust.