A Novel Spectral–Spatial Classification Method for Hyperspectral Image at Superpixel Level

A Novel Spectral–Spatial Classification Method for Hyperspectral Image at Superpixel Level
复制标题

一种新的超像素级高光谱图像光谱空间分类方法

DOI:
10.3390/app10020463
复制
发表时间:
2020-01
期刊:
影响因子:
--
通讯作者:
AN Na
AN Na
中科院分区:
--
文献类型:
--
作者:
XIE Fuding;LEI CunKuan;JIN Cui;AN Na

文献摘要

参考文献

相似文献

虽然超像素分割为高光谱图像提供了强大的工具。在超像素水平上对HSI进行分类仍然是一个具有挑战性的问题。超像素自适应大小和形状的特点。此外,这些特点。超像素和噪声像素的出现使得适当地测量图像变得困难。两个超像素之间的相似性。假设一个超像素内的像素属于。本文提出了一种新的光谱-空间HSI分类方法。超像素级方法(SSC-SL)。首先,提出了一种简单的线性迭代聚类算法。通过引入新的相似度和排名技术进行改进。改进的SLIC是专门为HSI设计的,可以直接将任意维度的HSI分割成超像素,而不需要。事先参考主成分分析。此外,超像素到超像素的相似性。是新推出的。定义的相似度与超像素的形状无关。减小了噪声像素对相似度的影响。最后,完成分类任务。通过根据最近的标记超像素标记每个未标记的超像素。在提议中。超像素级分类方案,每个超像素作为一个样本。这很明显。大大减少了需要分类的数据量。三个实高光谱的实验结果。数据集证明了所提出的光谱空间分类方法的优越性。几种比较先进的分类方法,在分类精度方面。
Although superpixel segmentation provides a powerful tool for hyperspectral image.(HSI) classification, it is still a challenging problem to classify an HSI at superpixel level because of.the characteristics of adaptive size and shape of superpixels. Furthermore, these characteristics of.superpixels along with the appearance of noisy pixels makes it difficult to appropriately measure the.similarity between two superpixels. Under the assumption that pixels within a superpixel belong to.the same class with a high probability, this paper proposes a novel spectral–spatial HSI classification.method at superpixel level (SSC-SL). Firstly, a simple linear iterative clustering (SLIC) algorithm is.improved by introducing a new similarity and a ranking technique. The improved SLIC, specifically.designed for HSI, can straightly segment HSI with arbitrary dimensionality into superpixels, without.consulting principal component analysis beforehand. In addition, a superpixel-to-superpixel similarity.is newly introduced. The defined similarity is independent of the shape of superpixel, and the.influence of noisy pixels on the similarity is weakened. Finally, the classification task is accomplished.by labeling each unlabeled superpixel according to the nearest labeled superpixel. In the proposed.superpixel-level classification scheme, each superpixel is regarded as a sample. This obviously.greatly reduces the data volume to be classified. The experimental results on three real hyperspectral.datasets demonstrate the superiority of the proposed spectral–spatial classification method over.several comparative state-of-the-art classification approaches, in terms of classification accuracy.
DOI: 10.1109/tgrs.2013.2275613
发表时间: 2014-06-01
影响因子: 8.2
作者:
Kang, Xudong;Li, Shutao;Benediktsson, Jon Atli
通讯作者: Benediktsson, Jon Atli
通过最小化奇异值部分和和超像素分割进行高光谱图像去噪
DOI: 10.1016/j.neucom.2018.11.039
发表时间: 2019-02-22
期刊: NEUROCOMPUTING
影响因子: 6
作者:
Liu, Yang;Shan, Caifeng;Cui, Rongmei
通讯作者: Cui, Rongmei
DOI: 10.3390/rs11131552
发表时间: 2019-06
期刊: Remote. Sens.
影响因子: --
作者:
Chunhua Dong;M. Naghedolfeizi;D. Aberra;Xiangyan Zeng
通讯作者: Chunhua Dong;M. Naghedolfeizi;D. Aberra;Xiangyan Zeng
DOI: 10.3390/rs9020139
发表时间: 2017-02
期刊: Remote. Sens.
影响因子: --
作者:
Shuzhen Zhang;Shutao Li;Wei Fu;Leyuan Fang
通讯作者: Shuzhen Zhang;Shutao Li;Wei Fu;Leyuan Fang
DOI: 10.1109/tgrs.2009.2037898
发表时间: 2010-02
影响因子: 8.2
作者:
F. Ratle;Gustau Camps-Valls;J. Weston
通讯作者: F. Ratle;Gustau Camps-Valls;J. Weston