摘要
为了提高水下目标识别的识别率,降低水下目标特征提取的代价,提出了基于二进制粒子群优化(Discrete Binary Parti- cle Swarm Optimization,BPSO)的水下目标特征选择算法,并结合k近邻分类算法,对三类实测水下目标数据进行了最优特征集的选择及分类实验。实验结果表明该特征选择方法能有效降低水下目标的特征维数,选择出利于分类的特征子集,提高了水下目标识别的分类效果。为了说明方法对于其他模式识别问题的效果,另外选择了UCI机器学习数据库中的四组标准数据进行仿真分析。
In this paper, a new feature selection method based on BPSO （Discrete Binary Particle Swarm Optimization） algorithm is proposed for improving classification accuracy and reducing feature dimensions of underwater acoustic targets. BPSO is used to select optimum feature subset, and k - NN is used to evaluate the performance of feature subset. Three different classes of underwater targets datasets and standard datasets from UCI machine learning repository are used in the experiment. The results show that the feature selection method based on BPSO is an effective technique for underwater acoustic targets＇ feature selection, and can enhance classification accuracy.
出处
《计算机仿真》
CSCD
2008年第1期196-199,共4页
Computer Simulation
关键词
粒子群优化算法
算法
特征选择
水下目标识别
Particle swarm optimization （PSO）
Algorithm
Feature selection
Underwater target recognition
