DBCR-YOLO: improved YOLOv5 based on double-sampling and broad-feature coordinate-attention residual module for water surface object detection

Abstract

Unmanned missions have become more and more popular in recent years. The related technologies of unmanned ground vehicles and unmanned aerial vehicles are growing rapidly, but research on unmanned surface vehicles (USVs) is rare. Water surface object detection algorithms play a crucial role in the field of USVs. However, achieving an object detection algorithm that balances speed and accuracy in the presence of interference is a difficult challenge. We proposed a network, DBCR-YOLO, that improved the detection accuracy while meeting real-time requirements. Based on YOLOv5, we added an additional detection head for detecting tiny objects. Then, we replaced the downsampling in YOLOv5's backbone network with the proposed double sampling mechanism to solve the problem that paying attention to the key features of objects cannot be done in the downsampling process of YOLOv5. Finally, we substituted the proposed BCR neck for YOLOv5's neck, thus improving the fusion of features between different scales based on fewer parameters and fewer calculations. We tested our network on the water surface object detection dataset. Compared with YOLOv5, DBCR-YOLO improved the detection accuracy by 3.4%. At the same time, DBCR-YOLO achieved the highest accuracy in comparison with other networks.

EBiDA-FPN: enhanced bi-directional attention feature pyramid network for object detection

“…Moreover, it also uses various dataaugmented strategies, such as Mosaic, copy-paste, and Mixup, which is one of the classic works in the field of object detection. Many subsequent works are based on it, including YOLOv5-R, 35 DBCR-YOLO, 36 and so on. FCOS 37 proposed the concept of anchor-free to eliminate the design of anchor boxes and the resulting problems, such as IoU computation while improving detection performance.…”