Abstract: In order to improve water quality evaluation of multi-spectral image accurately，this paper puts forward a model for water quality evaluation based on Support Vector Regression with parameters optimized by particle swarm optimization algorithms.The model uses high-resolution multi-spectral remote SPOT-5 data and the water quality field data，the parameters of Support Vector Regression are optimized by particle swarm optimization algorithms.First，the water quality parameters of Weihe River in Shaanxi Province are analyzed to choose four representative water quality parameters.Five methods are used to finish the correction of atmospheric radiation in remote sensing images.Then，the relevance between water quality parameters and remote sensing data is analyzed and retrieved.Finally，the proposed model is applied to the water quality evaluation of Weihe River in Shaanxi Province.The result of experiment shows the proposed method can give a better quality comprehensive evaluation，and can reflect the water quality of rivers accurately and objectively from the overall.It provides a new approach for evaluation of environment to inland rivers.

Key words: Support Vector Regression（SVR）, Particle Swarm Optimization（PSO） algorithm, parameter optimized, high-resolution remote sensing image, water quality evaluation

摘要： 为进一步提高多光谱图像水质反演的评价精度，提出了一种基于PSO优选参数的SVR水质评价方法。该模型利用高分辨率多光谱遥感SPOT-5 数据和水质实地监测数据，用粒子群优化算法对支持向量回归的参数进行了优化。首先，分析和筛选渭河陕西段水质实地监测数据，得到符合条件且具有代表性的四类水质变量。接着，使用五种大气校正方法对遥感影像进行大气辐射校正。然后，对各水质变量与遥感数据波段进行相关性分析和水质反演。最后，运用该模型以渭河水质监测数据为例进行了水质评价。实验结果表明，该方法可以较好地实现水质综合评价，能从整体上准确、客观地反映河流水质情况，为内陆河流环境评价提供了一种新方法。

关键词: 支持向量回归, 粒子群优化算法, 参数优选, 高分辨遥感影像, 水质评价