基于LBV变换的遥感影像多步骤分类法研究

面对与日俱增的遥感数据,如何快速准确地进行影像分类成为遥感领域亟待解决的问题。本文在前人的基础上提出了一种基于LBV变换的遥感影像多步骤分类方法。L:地物的总辐射水平,集中反映了裸地的信息;B:地物的可见-红外光辐射平衡,是地面水分状况和水体存在的一个良好指标;V:地物辐射随波段变化的方向和速度向量,能集中反映地面植被状况。多步骤分类法遵循由易到难的原则,能克服类别之间的互相影响,从而提高分类精度。文章结合LBV变换和多步骤分类法的优点对影像进行了分类,结果表明该方法简单易行,且能达到良好的分类精度。

Study on multi-step classification of remote sensing image based on LBV transformation

Facing the data of remote sensing increasing day by day,how to classify image quickly and accurately become the matter in remote sensing field.The paper brings forward a multi-ste Pclassification approach based on LBV transformation.L:general radiance level,it can reflect the information of bare soil.B:visible-infrared radiation balance,it's a good index of water.V:band radiance variation vector-direction and speed,it can reflect the condition of vegetation.Multi-step classification follows the rule from easy to difficult.It can overcome the impact of different objects.So this method can improve the accuracy of classification.This paper integrates the good qualities of LBV transformation and multi-step classification.The result shows that this approach is simple and easy to operate.Further more,good precision can be reached.