融和启发信息的蚁群特征选择方法

特征提取和选择是模式识别核心问题之一,它极大地影响着分类器的设计和性能,高维的特征选择更是一个NP难题。针对特征选择这一组合优化及多目标优化问题,本文提出了改进的融合启发信息ACO(Antcolony optimization)特征选择的新方法,该算法比不用启发信息的ACO方法能更好地找出代表问题空间的最优特征子集,降低分类系统的搜索空间,从而提高搜索效率。以航空纹理影像的特征选择和分类问题为例,利用原始蚂蚁算法和改进的蚂蚁算法选择的特征分别进行识别,结果证明该算法不仅能够比没有改进的蚂蚁找出有效特征集、降低图像特征空间维数、减少图像分类的工作量,而且提高了分类识别正确率。

A novel approach to feature selection based on Ant colony optimization algorithm Hybridization of heuristic information

In pattern recognition, feature extraction and selection has long been an important topic and has been studied by many authors because of its impact on the complexity of classifiers. Further more, feature selection in high dimension space is a NP problem. This paper present a novel approach to feature subset selection based on ACO(Ant colony optimization algorithm).the approach has the ability to accommodate multiple criteria such as accuracy and cost of classification into the process of feature selection and finds the effective feature subset for texture classification. A classifier based on minimum distance is described to label two types of texture images with feature subset selected by ACO and extracted by PCA (Principal Component Analysis )respectively. Experimental results illustrates that the algorithm can reduce feature dimension, speed the classification of image and improve the recognition rate compared to PCA.