SCIENCE@home

A Multiscale Superpixel-level Group Clustering Framework for Hyperspectral Band Selection

S. Jia,Yue Yuan,Nanying Li,Jianhui Liao,Qiang Huang,X. Jia,Meng Xu

Published 2022 in IEEE Transactions on Geoscience and Remote Sensing

ABSTRACT

Hyperspectral imagery (HSI) contains hundreds of bands, which provides a wealth of spectral information and enables better characterization of features. However, the excessive dimensionality also poses a dimensional disaster for subsequent processing. Fortunately, band selection gives a straightforward and effective way to pick out a subset of bands with rich information and low correlation. Although many hyperspectral band selection methods, especially clustering-based ones, have been proposed by researchers in recent years, the contextual information of adjacent bands and the spatial structural information of materials are not well investigated. Therefore, in this paper, a multiscale superpixel-level group clustering framework (MSGCF) is proposed for hyperspectral band selection. Different from previous, a new superpixel-level distance measure is elaborately utilized to group and cluster the spectral bands, which jointly considers the spectral context and spatial structure information. Concretely, to preserve the spatial structural information of HSI, multiple superpixel segmentation is firstly performed to generate superpixel maps in multi-scales, which enables complementarity of multiple superpixel segmentation algorithms and adaptation to diverse scales of land cover types. Secondly, the grouping and clustering paradigm is introduced to conduct the contextual information among bands. Here the maximum points of superpixel-level KL-ℓ1 distance of adjacent bands are adopted as partition points to separate bands into groups, which encourages adjacent bands with strong correlation to be divided into the same group. Thirdly, a superpixel-level fast density-based clustering method (SuFDPC) with superpixel-level ℓ2,1 distance is developed to select representative bands in every group. Finally, band selection results are achieved with a ranking-based voting strategy by concerning information entropy and frequency of occurrence in a unified scheme. A series of ablation analyses and experimental comparisons on four real HSI datasets have been conducted, as well as similarity comparisons for the selected bands. The experimental results consistently demonstrated the effectiveness of our MSGCF approach. The codes of this work will be available at http://jiasen.tech/papers/ for the sake of reproducibility.

PUBLICATION RECORD

  • Publication year

    2022

  • Venue

    IEEE Transactions on Geoscience and Remote Sensing

  • Publication date

    Unknown publication date

  • Fields of study

    Computer Science, Engineering, Environmental Science

  • Identifiers
    DOI 10.1109/TGRS.2022.3150361
  • External record

    Open on Semantic Scholar

  • Source metadata

    Semantic Scholar

CITATION MAP

EXTRACTION MAP

CLAIMS

  • No claims are published for this paper.

CONCEPTS

  • No concepts are published for this paper.

REFERENCES

Showing 1-97 of 97 references · Page 1 of 1

CITED BY

Showing 1-37 of 37 citing papers · Page 1 of 1