Abstract :
Hyperspectral Image (HSI) registers hundreds of spectral bands, whose intraclass variability and interclass similarity are resourceful information to be mined. Intraclass variability reflects the non-uniform and redundancy of the spatial and semantic features extracted from HSI. Interclass similarity represents the inherent relationship between adjacent features and snapshots. Existing models extract the superficial correlation representation for HSI to tackle the classification task but fail to embed the interclass and intraclass correlations due to these models intrinsic bottlenecks. Confronting the challenges of capturing interrelation for complex data in practice, we propose a Context-Aware Attentional Graph U-Net (CAGU) to improve these two modes of representation, which is more flexible in feature enhancement. In this method, attentional Graph U-Net is capable of extracting the intraclass embeddings within a non-Euclidean space by combining similar distributing feature vertices. The Gated Recurrent Unit (GRU) is another critical component of our model to capture the context-aware dynamic interclass embeddings. Extensive experiments demonstrate that our model can efficiently outperform state-of-the-art methods across-the-board on five wide-adopted public datasets, namely Pavia University, Indian Pines, Salinas Scene-show, Houston 2013 and Houston 2018, on a par with the same scale of model parameters.
CAGU Architecture :
Results :
Classification maps :
Innovation :
CAGU considers the intrinsic relationship in feature space, and promotes high cohesion of features through the graph network. CAGU can explore deeper spectral information in the image than the state-of-the-art methods, which enhances the cohesion of features and provides clear data for subsequent modules. Therefore, CAGU outperforms all the compared methods, and it can be universally employed in most models for feature enhancement, not restricted in U-Net. What’s more, GCN and GRU will not increase the parameters of the model due to the reduction in depth of U-Net.
Publish :
Journal : IEEE Geoscience and Remote Sensing Letters
Impact factor : 3.833
State : Accepted (28-Mar-2021)
Declaration :
When the paper is published, we will open source the code.
-- Moule Lin
