A2-FPN for semantic segmentation of fine-resolution remotely sensed images

The thriving development of earth observation technology makes more and more high-resolution remote-sensing images easy to obtain. However, caused by fine-resolution, the huge spatial and spectral complexity leads to the automation of semantic segmentation becoming a challenging task. Addressing such an issue represents an exciting research field, which paves the way for scene-level landscape pattern analysis and decision-making. To tackle this problem, we propose an approach for automatic land segmentation based on the Feature Pyramid Network (FPN). As a classic architecture, FPN can build a feature pyramid with high-level semantics throughout. However, intrinsic defects in feature extraction and fusion hinder FPN from further aggregating more discriminative features. Hence, we propose an Attention Aggregation Module (AAM) to enhance multiscale feature learning through attention-guided feature aggregation. Based on FPN and AAM, a novel framework named Attention Aggregation Feature Pyramid Network (A(2)-FPN) is developed for semantic segmentation of fine-resolution remotely sensed images. Extensive experiments conducted on four datasets demonstrate the effectiveness of our A(2)-FPN in segmentation accuracy. Code is available at .

Automated summary

The thriving development of earth observation technology has made it easier to obtain high-resolution remote-sensing images. However, the complexity caused by fine-resolution makes automated semantic segmentation a challenging task. To tackle this issue, researchers propose a novel framework called Attention Aggregation Feature Pyramid Network (A(2)-FPN) that uses the Feature Pyramid Network (FPN) and Attention Aggregation Module (AAM) to enhance multiscale feature learning. Extensive experiments demonstrate the effectiveness of A(2)-FPN in segmentation accuracy.