Meiping Song

TL;DR: The results and analysis of extensive real HSIC experiments demonstrate that the proposed light-weighted 2D-3D CNN network can effectively extract refined features and improve the classification accuracy.

TL;DR: Experimental results demonstrate BSNE-ICEM, which has advantages over support vector machine-based approaches in many aspects, such as easy implementation, fewer parameters to be used, and better false classification and precision rates.

TL;DR: A CNN system embedded with an extracted hashing feature is proposed for HSI classification that utilizes the semantic information of the HSI that achieves powerful distinguishing ability from different classes.

TL;DR: The feedback attention modules are developed for the first time to enhance the attention map with the semantic knowledge from the high-level layer of the dense model, and the spatial attention module is strengthened by considering multiscale spatial information.

TL;DR: This article proposed a novel method to exploit local spectral similarity through nonlocal spatial similarity, with the integration of local spatial consistency in a single framework that demonstrates the efficiency of the proposed method, with improvements over the other related nonlocal or local similarity-based methods.