In order to improve water quality retrievals of multi-spectral image accurately, a model for water quality remote retrieval is put forward based on Support Vector Regression (SVR) with parameters optimized by genetic algorithm. The model based on high-resolution multi-spectral remote SPOT-5 data and the water quality field data, uses CV to estimate the promote error. And parameters of SVR model are optimized by Genetic Algorithm. The global optimization of model parameters is achieved. Then, water quality is retrieved by the trained SVR. The proposed model is applied to the water quality retrievals of Weihe River in Shaanxi Province. The result of experiment shows that the developed model has more accuracy than that of the routine linear regression model，which provides a new approach for remote sensing monitoring of environment in inland rivers.