毕业论文--基于支持向量机的飞机目标识别.pdf

编号南京航空航天大学毕业论文题目基于支持向量机的飞机目标识别学生姓名熊钰城学号 070750904 学院民航学院专业民航机务工程班级 0708301 指导教师徐一鸣讲师二〇一二年六月南京航空航天大学本科毕业设计(论文)诚信承诺书本人郑重声明:所呈交的毕业设计(论文)(题目:基于支持向量机的飞机目标识别)是本人在导师的指导下独立进行研究所取得的成果。尽本人所知,除了毕业设计(论文)中特别加以标注引用的内容外,本毕业设计(论文)不包含任何其他个人或集体已经发表或撰写的成果作品。作者签名: 年月日(学号): 毕业设计(论文)报告纸 i 基于支持向量机的飞机目标识别摘要本文介绍当今的支持向量机技术及在飞机目标识别中的应用。在现代的战争中,目标识别成为一个决定胜负的重要因素,所以目标识别的技术成为现代军事的一个热门研究点,随着科技水平的不断提升与发展,侦查与追踪目标越来越难以准确的定位与识别,为了在军事行动中能更加快速准确的识别飞机目标,首先通过自动阈值分割,然后提取其灰度方差、不变矩特征和像素比组成特征向量,再用支持向量机作为分类器进行判别。支持向量机(SVM)是一种全新的机器学****方法,它建立在结构风险最小化准则的基础上且有较强的推广能力,适用于小样本训练情况,非常实用于飞机目标识别,同时还具备强大的处理能力,因此本文提出了一种基于支持向量机的飞机目标识别方法。关键词: 支持向量机(SVM),飞机目标识别,灰度方差。毕业设计(论文)报告纸 ii Aircrafttargetrecognition based on support vector machine Abstract This article describes support vector machine technology in today's aircraft target recognition. In modernwarfare, target identificationisan important factor to decide the warfare is winning or lose, so, thetarget recognition technology has e a hot point of the modern military research, rinsingwith the level of technology and development, detect and track the target more and more difficult to accurately positioning and recognition, in orderto quickly and accurately identify aircraft targets in military operations, the first should automatic thresholdsegmentation, and then extract the grayscale variance, the same moment feature and pixel composed of the eigenvectors, and then use the Support Vector Machine as a classifier to distinguish. Support Vector Machine(SVM) is a new machine learning method,the target scattering establish on the basis of the structural risk minimization criteria and it has strong generalization, its apply to the training of the small sample, it very practical aircraft target recognition, and it also has powerful processing capabilities, so this paper proposes a method of Aircrafttargetrecognition based on support vector machine. Keywords : support vector m