Gaussian Process Approach to Remote Sensing Image Classification
Gaussian Process Approach to Remote Sensing Image Classification
复制标题
DOI：
10.1109/tgrs.2009.2023983
复制
发表时间：
2010-01-01
影响因子：
8.2
通讯作者：
Melgani, Farid
中科院分区：
文献类型：
--
作者：
Bazi, Yakoub;Melgani, Farid
高斯过程(GP)代表了贝叶斯分类的一个强大而有趣的理论框架。尽管近年来声名显赫，但它们仍然是一种潜力尚不充分的方法。本文针对多源和高光谱遥感图像的分类问题，对GP方法进行了深入的研究。为此，我们探索了两种GP分类的解析逼近方法，即拉普拉斯方法和期望传播方法，这两种方法是用两种不同的协方差函数实现的，即平方指数协方差函数和神经网络协方差函数。此外，我们还分析了如何通过一种类似于信息向量机的快速稀疏近似方法，在不显著损失辨别能力的情况下大幅降低GP分类器(GPC)的计算负担。实验的目的也是为了测试GPC对训练样本数量和维度灾难的敏感度。总体而言，分类结果清楚地表明，广义预测控制可以与目前最先进的支持向量机分类器竞争。
Gaussian processes (GPs) represent a powerful and interesting theoretical framework for Bayesian classification. Despite having gained prominence in recent years, they remain an approach whose potentialities are not yet sufficiently known. In this paper, we propose a thorough investigation of the GP approach for classifying multisource and hyperspectral remote sensing images. To this end, we explore two analytical approximation methods for GP classification, namely, the Laplace and expectation-propagation methods, which are implemented with two different covariance functions, i.e., the squared exponential and neural-network covariance functions. Moreover, we analyze how the computational burden of GP classifiers (GPCs) can be drastically reduced without significant losses in terms of discrimination power through a fast sparse-approximation method like the informative vector machine. Experiments were designed aiming also at testing the sensitivity of GPCs to the number of training samples and to the curse of dimensionality. In general, the obtained classification results show clearly that the GPC can compete seriously with the state-of-the-art support vector machine classifier.