English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

The unique characteristics of hyperspectral images (HSI) undoubtedly pose significant categorization issues while providing a wealth of information. High redundancy and a lack of samples with labels are two main issues. Since convolutional neural networks were developed, several different network models have been utilized for hyperspectral image categorization. In this research, we introduce a multiscale piecewise spectral-spatial attention network (MPSSAN) that improves accuracy with fewer labelled samples. Concretely, we first randomly partition several groups to address the high dimensionality of hyperspectral data, and then the principal component analysis is utilized for dimension reduction. Following that, two multi-branch structures are employed to extract spectral and spatial features, and the features of multi-branch structures interact with each other to improve information flow. In addition, the designed double-scale attention mechanism could assign greater weight to crucial spectral-spatial features for further capturing effective HSI information. Experiments run on three datasets to test the model’s efficacy, and several popular deep-learning methods are selected for comparison experiments. The experimental findings demonstrate that the proposed MPSSAN enhances classification performance when compared to some state-of-the-art methods. Our method’s overall accuracy is 98.93% for the Indian Pines (IN) dataset and 99.68% for the Kennedy Space Center (KSC) dataset with 10% samples. On the University of Pavia (UP) dataset, 1% of training samples achieve an overall accuracy of 98.46%.

Hyperspectral image classification based on multiscale piecewise spectral-spatial attention network