基于模式识别的激光雷达遥感灰霾组分识别模型

Pattern Recognition Model for Haze Identification with Atmospheric Backscatter Lidars

提出了一种基于贝叶斯模式识别的激光雷达大气遥感灰霾组分识别的方法。介绍了灰霾组分模式识别模型的建立过程,并利用具体的贝叶斯判别函数作为灰霾粒子光学特征向量的选择依据对灰霾粒子进行识别分类。采用计算机仿真实现了该灰霾组分模式识别模型,并通过两种自验证方法检验了模型的正确性和稳定性。讨论了该模型对现有大气遥感激光雷达的适用性,凸显了偏振高光谱分辨率激光雷达(HSRL)的优势。

A pattern recognition model for haze identification with atmospheric backscatter lidars is proposed. The process of building the characteristics sample database for haze pattern recognition is described in detail. The classification of haze particles by using Bayesian discriminant function, as the selection basis of haze optical characteristics vector, is presented. Computer simulation for the proposed pattern recognition model of haze identification is carried out. Two self-calibration approaches are employed to check the validity and stability of the model. By analyzing the applicability of this model for atmospheric lidars, the advantage of polarized high spectral resolution lidar （HSRL） is highlighted.