摘要
针对高分辨率遥感图像舰船目标识别问题,提出了一种基于支持向量机的舰船目标分类方法。支持向量机(SVM)是一类新型机器学习方法,基于结构风险最小化归纳原则,具有出色的学习能力。与传统的方法相比,支持向量机不但结构简单,而且技术性能特别是泛化能力明显提高。该文简要介绍了有关统计学习理论和支持向量机算法,将支持向量机应用于遥感图像舰船目标识别,并同传统的舰船识别方法进行了相关的对比实验,实验结果说明本文提出的分类器在识别性能上明显优于其它传统分类器,具有更高的识别性能率。
Aiming at recognizing ship target in high resolution satellite images, a classification method based on Support Vector Machine(SVM) is proposed. SVM is a novel machine learning method, based on structural risk minimization principle and has excellent learning performance. Compared with many traditional methods, SVM is not only relatively simple in structure, but also shows better performances, especially better generalization ability. In this paper, Statistical Learning Theory and Support Vector Machine are briefly introduced, thenhow SVM is applied in ship target recognition in remote sensing images is recommended in detail. In the end, the experimental classification performance is compared with several traditional methods, results show that the proposed classifier has better classification rate than that of traditional classifiers.
出处
《计算机仿真》
CSCD
2006年第6期180-183,共4页
Computer Simulation
基金
中科院支撑技术预先研究项目(42201020501)
中科院知识创新方向性项目(kzcx0101)
关键词
支持向量机
统计学习理论
舰船目标分类
Support vector machine (SVM)
Statistical learning theory
Ship target recognition