提出一种高效可扩展的高光谱图像聚类方法，同时保留局部与非局部结构信息。
Scalable Context-Preserving Model-Aware Deep Clustering for Hyperspectral Images
- 基于基表示的一阶段框架，联合优化局部与非局部结构约束。
- 时间空间复杂度均为O(n)，支持大规模数据处理，准确率优于现有方法。
- 适合需要高效、高精度聚类的大规模高光谱图像分析任务。
子空间聚类已成为高光谱图像（HSI）无监督分析的主流方法。现有模型感知型深度子空间聚类方法多采用两阶段框架：先计算复杂度为O(n²)的自表示矩阵，再进行谱聚类。这类方法计算开销大，通常仅考虑局部或非局部空间结构约束，且约束难以有效指导整个聚类过程。本文提出一种可扩展的、上下文保持的深度聚类方法，基于基表示，联合捕捉局部与非局部结构以实现高效聚类。为保持局部结构（即子空间内的空间连续性），引入空间平滑约束，使聚类预测与其空间滤波版本对齐；为建模非局部结构（即光谱连续性），采用小簇级方案，在群体层面优化预测，促使光谱相似像素归属同一子空间。两个约束联合优化，相互增强。模型设计为单阶段架构，结构约束贯穿整个聚类过程。所提方法的时间与空间复杂度均为O(n)，适用于大规模高光谱数据。在真实数据集上的实验表明，该方法显著优于现有先进方法。代码已开源：https://github.com/lxlscut/SCDSC
原文摘要 · Abstract (English)
Subspace clustering has become widely adopted for the unsupervised analysis of hyperspectral images (HSIs). Recent model-aware deep subspace clustering methods often use a two-stage framework, involving the calculation of a self-representation matrix with complexity of O(n^2), followed by spectral clustering. However, these methods are computationally intensive, generally incorporating solely either local or non-local spatial structure constraints, and their structural constraints fall short of effectively supervising the entire clustering process. We propose a scalable, context-preserving deep clustering method based on basis representation, which jointly captures local and non-local structures for efficient HSI clustering. To preserve local structure (i.e., spatial continuity within subspaces), we introduce a spatial smoothness constraint that aligns clustering predictions with their spatially filtered versions. For non-local structure (i.e., spectral continuity), we employ a mini-cluster-based scheme that refines predictions at the group level, encouraging spectrally similar pixels to belong to the same subspace. Notably, these two constraints are jointly optimized to reinforce each other. Specifically, our model is designed as an one-stage approach in which the structural constraints are applied to the entire clustering process. The time and space complexity of our method is O(n), making it applicable to large-scale HSI data. Experiments on real-world datasets show that our method outperforms state-of-the-art techniques. Our code is available at: https://github.com/lxlscut/SCDSC
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。