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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Accurate understanding of the scene of ship ports is important in a broad range of military and civilian applications, such as maritime management, maritime safety, fisheries management, maritime situational awareness (MSA), and ocean traffic surveillance. Semantic segmentation, which implements pixel-level classification, can achieve an exhaustive analysis of ship ports. However, in the earlier time, the state-of-the-art methods of semantic segmentation were mostly based on the study of natural images. Subsequently, semantic segmentation has been gradually widely used in remote sensing, but still few of them has focused on the parsing of ship ports in remote sensing. In order to realize a detailed analysis of ship ports, a novel framework (called Feature Attention Network) is proposed for the accurate segmentation of multiple targets in a ship port in this paper. In this framework, a multi-label classification auxiliary network is first designed to solve the problem of confused multiple prediction for one target by capturing more global context information. Then, an attention model is introduced to solve the problem of error segmentation between similar targets with different labels. Finally, a feature aggregation model is presented to obtain more contextual information. In addition, we construct a data set for the semantic segmentation of ship ports (called HRSC2016-SP) by labeling the HRSC2016 data set to evaluate our proposed framework. Our approach has achieved a state-of-the-art result (82.16% mIoU) on the test set of HRSC2016-SP.

FA-Net: feature attention network for semantic segmentation of ship port