Sailboat Detection Based on Automated Search Attention Mechanism and Deep Learning Models

The development of deep learning-based visual object recognition has achived great progress in recent years. Deep learning models based on attention mechanisms are able to further improve the ability to detect the regions of interest (ROI), but creating an appropriate module is a difficult task. Therefore, in this paper, we propose an automated design scheme based on neural architecture search (NAS) to migrate the attention mechanism for visual object detection and obtain better results of sailboat detection by using both public datasets and our own collected datasets. We verify the effectiveness of our proffered method and evaluate the performance compared with other algorithms. The obtained results have demonstrated better robustness in terms of generalization than other deep learning models.