cs.CVNov 12, 2026

MALNISI-Infused Graph Construction and Weighted Contrastive Loss based GAT for Unsupervised Representation Learning in HSI

Authors: Soumi Sarkar, Subhadip Boral, Sarbani Palit, Ashish Ghosh

Organizations: Machine Intelligence Unit, Indian Statistical Institute, Kolkata, West Bengal 700108, India · Siksha ’O’ Anusandhan, Bhubaneswar, Odisha 751030, India · Computer Vision and Pattern Recognition Unit, Indian Statistical Institute, Kolkata, West Bengal 700108, India · International Institute of Information Technology, Bhubaneswar, Odisha 751003, India

Abstract

Hyperspectral images (HSI) possess discriminating spectral characteristics with respect to land utilization due to high-dimensional information. Recently, unsupervised representation learning using graph neural networks has unveiled these utilization-specific complex spectral relations. This article has three main contributions. First, it proposes an index, the Mean Absolute Local Neighborhood Inverse Spectral Index (MALNISI), which measures the similarity between two pixels based on their neighborhood information. The index is used to construct a weighted adjacency matrix, giving the HSI a graph structure. Second, it proposes an upgraded Graph Attention Network that learns compact latent representations of hyperspectral data from the constructed graph. In this network, an enhanced multi-head attention mechanism incorporates the MALNISI weights as an auxiliary feature to emphasize learned attention between neighboring pixels. To alleviate over-smoothing and preserve fine-grained spectral details, the architecture integrates residual connections, skip projections, and feature normalization, which together enhance representation capability, stabilize learning, and accelerate convergence. Third, it proposes a weighted contrastive loss function that incorporates neighborhood-aware MALNISI between connected pixels, encouraging strongly connected pixels in the high-dimensional space to remain close in the latent space, and weakly connected pixels to stay apart. The resulting discriminative low-dimensional embeddings are then clustered using spectral clustering. Experimental analysis on six benchmark HSI datasets demonstrates the efficiency and necessity of the three aforementioned contributions, and the comparative study highlights superior performance over fourteen state-of-the-art approaches.

Figures & tables

Explore similar work

CardsList
  1. Deep Spatially-Regularized and Superpixel-Based Diffusion Learning for Unsupervised Hyperspectral Image Clustering

    Apr 14, 2026Vutichart Buranasiri, James M. MurphySuperpixel SegmentationClustering

  2. Manifold Prior Guided Deep Unfolding for Hyperspectral Image Reconstruction

    Oct 20, 2026Shuqi Liu, Yingkai Zhang, Lin Gu +2Hyperspectral Image ReconstructionDeep Unfolding

  3. DAPGNet: Dynamic Adaptive Physics-Guided Graph Diffusion Network for Hyperspectral Image Classification

    Jul 16, 2026Pengkun Wang, Weijia Cao, Ning Wang +1Graph Structure LearningHyperspectral Imaging