运用禁忌搜索算法进行高光谱影像降维的研究

Research on Tabu Search Algorithm in Hyperspectral Dimensionality Reduction

本文在分析了传统降维方法所面临问题的基础上，将禁忌搜索算法引入到高光谱影像的特征选择研究，指出由于禁忌搜索算法所具有的良好全局寻优能力，因而在该类影像的降维研究中有着广阔的应用前景。针对高维光谱数据的特点，讲述了算法运行过程中需注意的若干关键问题。实验表明，将禁忌搜索算法获取的波段进行高光谱影像分类，在求解的时间上和分类结果的精度都可达到令人满意的效果。

Based on analyzing the limitations of the traditional dimensionality reduction methods, this paper introduces the Tabu Search algorithm into hyperspectral feature selection. It is pointed out that due to the advantage of solving the global optimal problems, Tabu Search algorithm will be widely used in hyperspectral feature reduction. Moreover, some key problems during processing have been discussed. Experimental results demonstrate that Tabu Search is very helpful for obtaining good classification performance by using selected bands for classification.