Hyperspectral Classification with Swarm Intelligence Optimization Algorithms
Hyperspectral Classification with Swarm Intelligence Optimization Algorithms
复制标题
DOI：
10.1166/sl.2012.2638
复制
发表时间：
2012-12
期刊：
影响因子：
--
通讯作者：
Sheng Ding;Q. Qin;Li Chen;Hong Zhang
中科院分区：
文献类型：
--
作者：
Sheng Ding;Q. Qin;Li Chen;Hong Zhang
本文提出了一种新的高光谱图像分类系统，解决了高光谱遥感图像的波段选择和支持向量机参数优化问题。我们提出了一种基于粒子群优化（PSO，一种基于群体智能的方法）的元启发式优化分类系统，以提高SVM分类器的泛化性能。为此，我们通过搜索调整其判别函数的参数的最佳值来优化SVM分类器设计，并通过寻找馈送分类器的特征的最佳子集来优化上游。提出的PSO-SVM算法进行选择最佳的鉴别功能和合适的SVM参数的高光谱遥感图像的同时。最后通过粒子群优化算法优化SVM分类器的性能。该方法的有效性进行了评估，通过比较它与其他高光谱技术在文献中存在。在基准高光谱数据集上的实验结果证实了该方法的有效性。
This paper proposes the use of a new classification system for hyperspectral images, addresses the problem of band selection for hyperspectral remote sensing image and SVM parameter optimization. We propose a meta-heuristic optimization classification system based on particle swarm optimization (PSO, a swarm intelligence-based methodology) to improve the generalization performance of the SVM classifier. For this purpose, we have optimized the SVM classifier design by searching for the best value of the parameters that tune its discriminant function, and upstream by looking for the best subset of features that feed the classifier. The proposed PSO-SVM algorithm is performed to select the best discriminant features and appropriate SVM parameters for hyperspectral remote sensing imagery simultaneously. The performance of the SVM classifier is finally optimized through PSO. The effectiveness of the proposed method is evaluated by comparing it with other hyperspectral techniques existing in the literature. Experimental results on a benchmark hyperspectral dataset and confirmed the effectiveness of the proposed technique.