• 主题
高级搜索  >
历史搜索 清空历史
首页> 外文会议>Pixels, Objects, Intelligence: GEOgraphic Object Based Image Analysis for the 21st Century >QUANTITATIVE COMPARISON OF SEGMENTATION RESULTS FROM IKONOS IMAGES SHARPENED BY DIFFERENT FUSION AND INTERPOLATION TECHNIQUES
【24h】

QUANTITATIVE COMPARISON OF SEGMENTATION RESULTS FROM IKONOS IMAGES SHARPENED BY DIFFERENT FUSION AND INTERPOLATION TECHNIQUES

机译：由不同融合和插值技术锐化的IKONOS图像的分割结果的定量比较

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

摘要

The image sharpening and segmentation procedures have great importance on the classification success. For this reason this work proposes to evaluate segmentation results obtained from IKONOS images sharpened by different image fusion methods and interpolation techniques. The images were processed by different image fusion methods based on Principal Components, Grand-Schmidt and Wavelet transform. The resulting images are evaluated in terms of spectral information preservation. All hybrid images were segmented using algorithms based on region growing implemented in the SPRING 4.3 and eCognition 5.0 systems. The same segmentation parameters were used in all experiments. Segments obtained from thirty-two test-areas on an urban landscape were quantitatively compared by statistic t-tests. The results showed that the standardization of segmentation parameters and classification methodologies are not possible without taking into account the fusion and interpolation techniques used in the sharpening process.
机译：图像锐化和分割程序对分类成功非常重要。因此，这项工作建议评估从不同图像融合方法和插值技术锐化的Ikonos图像获得的分段结果。通过基于主组件，Grand-Schmidt和小波变换的不同图像融合方法处理图像。在光谱信息保存方面评估所得到的图像。使用基于弹簧4.3和Ecognition 5.0系统实现的区域生长的算法进行分段所有混合图像。所有实验中使用了相同的分段参数。通过统计T检验比较了从城市景观的三十二个测试区域获得的段。结果表明，不考虑锐化过程中使用的融合和插值技术，不可能进行分割参数和分类方法的标准化。

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