Efficient Superpixel Generation for Polarimetric SAR Images with Cross-Iteration and Hexagonal Initialization

被引:5
作者
Li, Meilin [1 ]
Zou, Huanxin [1 ]
Qin, Xianxiang [2 ]
Dong, Zhen [1 ]
Sun, Li [1 ]
Wei, Juan [1 ]
机构
[1] Natl Univ Def Technol, Coll Elect Sci & Technol, Changsha 410073, Peoples R China
[2] Air Force Engn Univ, Coll Informat & Nav, Xian 710077, Peoples R China
基金
中国国家自然科学基金;
关键词
polarimetric synthetic aperture radar (PolSAR); superpixel; cross-iteration; hexagonal initialization; geodesic distance; revised Wishart distance; CLASSIFICATION; SEGMENTATION; DISTANCE; NETWORK; DRIVEN;
D O I
10.3390/rs14122914
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Clustering-based methods of polarimetric synthetic aperture radar (PolSAR) image superpixel generation are popular due to their feasibility and parameter controllability. However, these methods pay more attention to improving boundary adherence and are usually time-consuming to generate satisfactory superpixels. To address this issue, a novel cross-iteration strategy is proposed to integrate various advantages of different distances with higher computational efficiency for the first time. Therefore, the revised Wishart distance (RWD), which has better boundary adherence but is time-consuming, is first integrated with the geodesic distance (GD), which has higher efficiency and more regular shape, to form a comprehensive similarity measure via the cross-iteration strategy. This similarity measure is then utilized alternately in the local clustering process according to the difference between two consecutive ratios of the current number of unstable pixels to the total number of unstable pixels, to achieve a lower computational burden and competitive accuracy for superpixel generation. Furthermore, hexagonal initialization is adopted to further reduce the complexity of searching pixels for relabelling in the local regions. Extensive experiments conducted on the AIRSAR, RADARSAT-2 and simulated data sets demonstrate that the proposed method exhibits higher computational efficiency and a more regular shape, resulting in a smooth representation of land cover in homogeneous regions and better-preserved details in heterogeneous regions.
引用
收藏
页数:30
相关论文
共 39 条
  • [1] SLIC Superpixels Compared to State-of-the-Art Superpixel Methods
    Achanta, Radhakrishna
    Shaji, Appu
    Smith, Kevin
    Lucchi, Aurelien
    Fua, Pascal
    Suesstrunk, Sabine
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2012, 34 (11) : 2274 - 2281
  • [2] A Graph-Based Semisupervised Deep Learning Model for PolSAR Image Classification
    Bi, Haixia
    Sun, Jian
    Xu, Zongben
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2019, 57 (04): : 2116 - 2132
  • [3] Comaniciu D., 1999, Proceedings of the Seventh IEEE International Conference on Computer Vision, P1197, DOI 10.1109/ICCV.1999.790416
  • [4] Polarimetric Contextual Classification of PolSAR Images Using Sparse Representation and Superpixels
    Feng, Jilan
    Cao, Zongjie
    Pi, Yiming
    [J]. REMOTE SENSING, 2014, 6 (08) : 7158 - 7181
  • [5] Superpixel-Driven Optimized Wishart Network for Fast PolSAR Image Classification Using Global k-Means Algorithm
    Gadhiya, Tushar
    Roy, Anil K.
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2020, 58 (01): : 97 - 109
  • [6] Decomposition-Feature-Iterative-Clustering-Based Superpixel Segmentation for PolSAR Image Classification
    Ho, Biao
    Yang, Chen
    Ren, Bo
    Jiao, Licheng
    [J]. IEEE GEOSCIENCE AND REMOTE SENSING LETTERS, 2018, 15 (08) : 1239 - 1243
  • [7] Superpixel Segmentation of Polarimetric Synthetic Aperture Radar (SAR) Images Based on Generalized Mean Shift
    Lang, Fengkai
    Yang, Jie
    Yan, Shiyong
    Qin, Fachao
    [J]. REMOTE SENSING, 2018, 10 (10):
  • [8] Mean-Shift-Based Speckle Filtering of Polarimetric SAR Data
    Lang, Fengkai
    Yang, Jie
    Li, Deren
    Shi, Lei
    Wei, Jujie
    [J]. IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2014, 52 (07): : 4440 - 4454
  • [9] TurboPixels: Fast Superpixels Using Geometric Flows
    Levinshtein, Alex
    Stere, Adrian
    Kutulakos, Kiriakos N.
    Fleet, David J.
    Dickinson, Sven J.
    Siddiqi, Kaleem
    [J]. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2009, 31 (12) : 2290 - 2297
  • [10] Li M., 2020, ISPRS ARCH, V43, P1225, DOI [10.5194/isprs-archives-XLIII-B2-2020-1225-2020, DOI 10.5194/ISPRS-ARCHIVES-XLIII-B2-2020-1225-2020]