Gaussian Radial Kernel Function Research Articles

A dataset of mangrove forest changes on Hainan Island based on GF-2 data during 2015–2019

Mangrove forests are wetland ecosystems in the coastal zone with important ecological and eco-economic service values. At the same time, mangrove forests are vulnerable ecosystems, being under threat from both natural and anthropogenic forces, so mangrove monitoring is a permanent task. The development of remote sensing technology has provided an efficient and convenient means for mangrove monitoring. In this study, we adopted a support vector machine (SVM) classification method with Gaussian radial kernel function, penalty factor of 100 and Gamma function 0f 0.022 to obtain a dataset of mangrove forest changes on Hainan Island from 2015 to 2019 by using Gaofen-2 (GF-2) data in 2015, 2017 and 2019 in combination with field survey data. The overall classification accuracy of this dataset is more than 99%, and the data volume is 58.7 MB. It can be used as the basic data for the analysis of spatial and temporal changes of mangrove forests. It can also provide decision-making support for the restoration, protection and management of mangrove wetland ecosystems, and facilitate the ecological environment supervision in Hainan Province.

On the condition number anomaly of Gaussian correlation matrices

Spatial correlation matrices appear in a large variety of applications. For example, they are an essential component of spatial Gaussian processes, also known as spatial linear models or Kriging estimators, which are powerful and well-established tools for a multitude of engineering applications such as the design and analysis of computer experiments, geostatistical problems and meteorological tasks.In radial basis function interpolation, Gaussian correlation matrices arise frequently as interpolation matrices from the Gaussian radial kernel function. In the field of data assimilation in numerical weather prediction, such matrices arise as background error covariances.Over the past thirty years, it was observed by several authors from several fields that the Gaussian correlation model is exceptionally prone to suffer from ill-conditioning, but a quantitative theoretical explanation for this anomaly was lacking. In this paper, a proof for the special position of the Gaussian correlation matrix is given. The theoretical findings are illustrated by numerical experiment.