• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Asian conference on remote sensing;ACRS >A Support Vector Machine Approach on Object Based Image Analysis for Feature Extraction from High Resolution Images
【24h】

A Support Vector Machine Approach on Object Based Image Analysis for Feature Extraction from High Resolution Images

机译：基于支持对象的图像分析的支持向量机从高分辨率图像中提取特征

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Satellite images are the most important available data sources for generation and updating of available maps. They have highly improved in terms of spatial, spectral and temporal resolutions and by the sheer volume of collected images, the necessity of simplification of automation in feature extraction. Road data play a key role in urban planning, traffic management, military applications, and vehicle navigation as well as for decision making in numerous applications. The faster updation of road infrastructure is a need because the technology has brought map in the hands of people in the form of mobile phones and tablets. Road detection is one of the major issues of the road infrastructure extraction. Its accuracy depends on the type of methodology used. An attempt is made here to analyse the first order, the co-occurrence texture features and image transforms useful for discriminating roads from other features specially the buildings. The identified dataset forms high dimension feature space and the Support Vector Machine is a theoretically superior machine learning methodology with great results in classification of high dimensional datasets. In the past, SVMs have been tested and evaluated only as pixel-based image classifiers. Moving from pixel-based techniques towards object-based representation, the dimensions of remote sensing imagery feature space increases significantly. An SVM approach for classification was followed, based on primitive image objects produces by a multi-resolution segmentation algorithm. The SVM procedure produced the final object classification results which were compared to the Nearest Neighbor classifier results and were found to give better results in OBIA domain.
机译：卫星图像是用于生成和更新可用地图的最重要的可用数据源。它们在空间，光谱和时间分辨率方面以及通过收集的图像数量庞大而得到了极大的改进，简化了特征提取中自动化的必要性。道路数据在城市规划，交通管理，军事应用，车辆导航以及众多应用的决策中起着关键作用。需要快速更新道路基础设施，因为该技术以手机和平板电脑的形式将地图带入了人们的手中。道路检测是道路基础设施提取的主要问题之一。其准确性取决于所使用方法的类型。此处尝试分析一阶，共现纹理特征和图像变换，这些特征可用于区分道路和其他特征（尤其是建筑物）。识别出的数据集形成高维特征空间，并且支持向量机是一种理论上优越的机器学习方法，在高维数据集的分类中取得了显著成果。过去，仅对SVM作为基于像素的图像分类器进行过测试和评估。从基于像素的技术转向基于对象的表示，遥感影像特征空间的尺寸显着增加。遵循基于多分辨率分割算法产生的原始图像对象的SVM分类方法。 SVM过程产生了最终的对象分类结果，将其与最近邻分类器结果进行比较，发现在OBIA域中可以提供更好的结果。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号