摘要
针对水下目标信号复杂、样本获取难度大、样本数目偏少的问题,提出了多类支持向量机集成算法,并且以此算法为核心构建了水下目标自动识别系统。通过对4类实测水下目标的识别实验,证明了所提出的水下目标自动识别系统可以用于水下目标识别,并且具有很好的推广能力。
A new ensemble-based approach is proposed for underwater target recognition that uses support vector machine ensembles （ SVME）. Four different classes of underwater target datasets are used in the experiment. The results show that the proposed SVME algorithm not only can improve the classification accuracy, but also generalize well.
出处
《测控技术》
CSCD
2006年第12期14-16,共3页
Measurement & Control Technology
关键词
支持向量机集成
水下目标
自动目标识别
support vector machine ensembles
underwater target
automatic target recognition
