空间光谱差比参量及其模拟应用

Spatial-spectral Difference Ratio Parameter and Its Simulation Application

针对自然界中广泛存在的光谱特征随空间有规律变化的一类地物在混合像元背景地物的干扰下难以有效识别的问题,提出了基于空间光谱差比参量的识别方法。通过理论分析与公式推导,首先对空间光谱差比参量的概念进行了定义,然后基于图像模拟.通过算法设计应用该参量进行目标识别。结果表明,该参量对于不同目标背景干扰下该类地物的识别具有科学性及有效性,模拟图像的分类识别精度达到了99.8%,随着影像空间分辨率与光谱分辨率的逐步提高,具有一定的应用前景。

Aiming at the problem that it is hard to identify the objects whose spectral feature regularly changes following the space changes by traditional methods,especially under the different sorts of interference in the nature,a new target identification method based on spatial-spectral difference ratio(Vss)parameter was developed.Firstly,we defined the concept of the Vss parameter through theoretical analysis and formula derivation.Secondly,based on image simulation,we used Vssparameter to identify the targets via the designed algorithm.The results show that it is scientific and effective to using the Vssparameter to identify targets which are disturbed by different background.The classification accuracy reached 99.8%in simulation image experiments.It will have certain application prospects with the gradual improvement of the image spatial resolution and spectral resolution.