我们将最近邻相关伊辛模型(INNC)应用于规则网格上空间相关数据的插值问题。这些关联是通过“Ising spins”之间的短程相互作用捕捉到的。INNC算法可用于标签数据(分类)以及离散和连续实值数据(回归)。在回归问题中,INNC通过用户指定数量的类来逼近连续变量。INNC根据观测数据(部分样本),使用蒙特卡罗模拟预测未测量点的类同一性。该算法局部考虑样本值,全局目标是最小化部分样本的能量度量与整个网格的能量度量之间的偏差。INNC是非参数的,因此适用于非高斯数据。与一些标准方法相比,该方法在插值精度和计算效率方面极具竞争力。因此,它为填补网格数据(如卫星图像)中的空白提供了一个有用的工具。
We apply the Ising model with nearest-neighbor correlations (INNC) in the problem of interpolation of spatially correlated data on regular grids. The correlations are captured by short-range interactions between “Ising spins”. The INNC algorithm can be used with label data (classification) as well as discrete and continuous, real-valued data (regression). In the regression problem, INNC approximates continuous variables by means of a user-specified number of classes. INNC predicts the class identity at unmeasured points using Monte Carlo simulation conditioned on the observed data (partial sample). The algorithm respects locally the sample values and globally aims to minimize the deviation between an energy measure of the partial sample and that of the entire grid. INNC is non-parametric and thus suitable for non-Gaussian data. The method is found to be very competitive with respect to interpolation accuracy and computational efficiency compared to some standard methods. Thus, it provides a useful tool for filling gaps in gridded data such as satellite images.
参考文献:
Žukovič, M.; Hristopulos, D.T., Ising Model for Interpolation of Spatial Data on Regular Grids.Entropy 2021, 23, 1270. https://doi.org/10.3390/e23101270
下载地址:
https://url92.ctfile.com/f/1850492-518390042-490e91
(访问密码:3660)
更多精彩文章请关注公众号: