摘要
为了提高遥感图像分类结果的正确率,提出了一种基于改进支持向量机的测绘遥感图像分类方法。采用FSA算法对SVM的核宽度和惩罚系数进行优化,得到FSA-SVM算法,采用实验数据进行仿真分析,并与其他遥感图像分类方法对比。结果表明,FSA-SVM算法分类结果的正确率和运算时间分别为98.34%和14.23s,分类效果明显优于其他算法,验证了所提测绘遥感图像分类方法的有效性。
In order to improve the accuracy of remote sensing image classification results,a remote sensing image classification method based on improved support vector machine is proposed.The FSA algorithm is used to optimize the kernel width and penalty coefficient of SVM,and the FSA-SVM algorithm is obtained.Experimental data is used for simulation analysis and compared with other remote sensing image classification methods.The results show that the accuracy and operation time of the classification results obtained by FSA-SVM algorithm are 98.34%and 14.23 seconds,respectively.The classification effect is significantly better than other algorithms,verifying the effectiveness of the proposed classification method forsurveying and mapping remote sensing images.
作者
王江林
WANG Jianglin(Guangzhou Southern Surveying and Mapping Technology Co.,Ltd.,Guangzhou 510663,China)
出处
《江西测绘》
2024年第1期12-15,共4页
JIANGXI CEHUI
基金
陆海一体化北斗三号精密定位服务与示范应用项目(项目编号:2023B1111050013)成果之一
关键词
遥感图像
分类
支持向量机
鱼群算法
测绘
Remote Sensing Image
Classification
Support Vector Machine
Fish School Algorithm
Surveying and Mapping
