首页> 外文会议>Workshop on Hyperspectral Image and Signal Processing: Evolution in Remote Sensing >On spatial regularization for semisupervised hyperspectral image segmentation using hybrid extreme rotation forest
【24h】

On spatial regularization for semisupervised hyperspectral image segmentation using hybrid extreme rotation forest

机译:混合极端旋转林的半培育高光谱图像分割的空间正则化

获取原文

摘要

This paper proposes a new semisupervised segmentation algorithm for hyperspectral image segmentation. The algorithm steps are as follows: 1) supervised training an initial classifier from a small balanced training set, 2) clustering of the image pixels, by a k-means algorithm 3) adding unlabeled pixels to the original training data set according to the spatial neighborhood and the cluster membership, 4) supervised training of the classifier with the enriched training data set, 6) classification of the hyperspectral image 4) spatial regularization of classification results consisting in selecting the most frequent class in each pixel neighborhood. In this work we use the Hybrid Extreme Rotation Forest (HERF) which has been successfully applied in medical image domain. Results on two well known benchmarking hyperspectral images improve over state of the art algorithm results.
机译:本文提出了一种新的超光图像分割的新半体育分割算法。该算法步骤如下:1)监督训练训练从一个小平衡训练集的初始分类器,2)通过K-means算法3的图像像素的聚类,通过k-means算法3增加了未标记的像素根据空间的原始训练数据集。邻域和群集成员资格,4)对分类器的培训具有丰富的训练数据集,6)高光谱图像的分类4)分类结果的空间正则化,其包括在每个像素邻域中选择最常见的类。在这项工作中,我们使用已成功应用于医学图像域的混合极端旋转森林(HERF)。结果两个众所周知的基准测试高光谱图像改善了最先进的算法结果。

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