For Water Quality Monitoring Of Key Techniques Of Remote Sensing Image Inversion

Nowadays, the global attaches great importance to environmental problem. It is improving the quality of environment and preventing it from being polluted that really matters. However, as an important branch of environmental pollution, water pollution has a terrible effect on people’s lives. How to deal with and do a wider and real-time monitoring of water pollution seems to become a hot issue both at home and abroad. With the development of the space science and technology, the highlighted satellite remote sensing image stands out, with provides a new way to monitor the quality of water pollution in a usual way conveniently; What’s more, through combining both aspects effectively, monitoring the quality of water will become more real-time, intelligent and visual.It start by the unique characteristic of spectral data and the advantage of remote sensing technology, and fully considering the spectral curve and the domestic high resolution multispectral satellite remote sensing image’s spectral characteristics, used for spectral data preprocessing and feature analysis, also on this basis, combining with the modeling ideas of water quality inversion method to fullfill the modeling construction. By obtaining the Pearson correlation between normalized, first-order differential their respective reflectance values and the real density of suspended solids, positioning the remote sensing spectrum range of GF-1, it determine sensitive band or band combination as the final modeling variables. It proves the experience of that the multiband ratio method can improve the the single band correlation coefficient. For the research object in Longquan Lake, making unique geometric correction, FLASSH atmospheric correction and water extraction. Confirmed that the single band method to extract water mask is more applicable for Longquan Lake.The key problem of the research of water quality monitoring is to solve building and application of water quality inversion model. From the perspective of improving water quality inversion model accurately, it carried out a study of based on the semi-empirical method to build the water quality inversion model and hybrid genetic algorithm in the application of the water quality inversion, and examined the unified indicators suggest that the hybrid genetic algorithm from optimization direction can indeed improve the inversion accuracy. Data also showed the Genetic Ant algorithm through the complementary advantages can make the solving time and efficiency to achieve optimization. From the perspective of engineering application, based on the professional software IDL for Romote Sensing Image Processing and visual interface processing mechanism, complete the water quality inversion for GF-1 of the high resolution and wide field monitoring, and thematic mapping. To intuitively show the effect of that the concentration distributions of water quality parameters, and clearly indicate the level of that the degree of waters eutrophication.In order to validate the user-friendly of water quality inversion system and the precision of inversion model, test procedures designed to show that water quality inversion system each function meet the operation requirements, and contrast experiment designed to show that the hybrid genetic algorithm proposed in this paper than the conventional inversion is enough to improve the precision of inversion results. In the end this article can provide strong support for the environmental pollution of water quality monitoring of Chengdu research project.