国产欧美精品一区二区,中文字幕专区在线亚洲,国产精品美女网站在线观看,艾秋果冻传媒2021精品,在线免费一区二区,久久久久久青草大香综合精品,日韩美aaa特级毛片,欧美成人精品午夜免费影视

Sudden water pollution accidents in industrial and mining enterprises have obvious suspended solids or obvious color changes, and most pollution can be judged intuitively through the visual way of surface water, based on the monitoring demand analysis of sudden water pollution accidents, a water surface pollution monitoring system based on industrial vision is studied in this paper. Aiming at the problem that continuous multi frame image information is affected by water flow velocity, particle suspended solids and system noise, a random disturbance signal filtering method and a multi-level image information adjustment method based on color information evaluation are proposed. Then, the pollution analysis model is established by using optimized neural network, and a monitoring device is developed to analyze the water image in real time to realize the rapid judgment of pollution state. After the system is used in the emergency station of a smelting enterprise, it reduces the labor intensity, shortens the judgment time, improves the judgment accuracy, reduces the incidence of environmental protection accidents, reduces the number of post personnel and reduces the cost, and has achieved remarkable application results.