广义高斯模型阈值的南极冰盖冻融探测

Detecting the Antarctic melting zones based on generalized Gaussian model threshold

在微波散射计的简单冰盖融化物理模型的基础上,提出了不依赖于实地观测数据的自动阈值分割的南极冰盖冻融探测的新算法,即利用广义高斯模型对通过微波散射计简单冰盖融化物理模型变换得到的数据做直方图统计,得到干湿雪划分的最优阈值,从而得到干湿雪的冻融分布图.这种算法不但解决了物理模型算法需依赖实测冰盖融化信息才能进行冰盖冻融探测的问题,而且能自动选择较多的样本,提高了冰盖冻融探测的计算效率、实用性和可操作性.算法也能被应用到其他地区,为全球的冰盖冻融探测提供了方法学的支持和补充.

On the basis of the simple snowmelt physical model for microwave scatterometer,a newautomatic threshold segmentation algorithm for detecting Antarctic ice-sheet melting zones was proposed,which did not depend on field observation. That is the histogram with the data of the physical model by using generalized Gaussian model to automatically get Antarctic melting zone distribution. The algorithm improves the computational efficiency,usability and operability of the melting zones,because the algorithm does not rely on the actual melt information and can automatically select more samples. The algorithm can also be applied to other regions,providing methodological support and supplement for the global melting zone detection.