摘要
为促进海洋资源开发,提高海洋开发能力,本文对水下目标分类识别方法进行研究。首先,对水下目标分类方法进行概述,介绍较为常用的方法。然后,提出K-means与SVM结合的水下目标分类方法。该方法利用S变换进行图像预处理,提取不同分辨率下的不同特征作为分类的特征向量,通过K-means与SVM结合的分类识别方法进行分类。实验结果表明,该方法具有较高的识别率。
To promote the development of marine resources, enhance the ability of marine development,underwater target classification and recognition methods were studied in this paper. First,this paper overviewed the underwater target classification,introduced methods used commonly. Then,K-means and SVM combination classification of underwater targets was proposed,the method using S-transform image preprocessing,extraction of the different characteristics of different resolutions as feature vector,and classified by K-means and SVM classification identification classification. Experimental results show that this method had a high recognition rate.
出处
《舰船科学技术》
北大核心
2015年第2期204-207,共4页
Ship Science and Technology
基金
2014年河南省教育厅科学技术重点研究资助项目(14A520045)
2013年河南省教育厅科学技术研究重点资助项目(13A520221)
作者简介
戴冬（1978-），女，硕士，讲师，研究方向为智能算法等。
