A Context-Enhanced Network for Underwater Pipeline Edge Detection Based on Keypoint Representation

Underwater pipeline edge detection using underwater imaging sensors is the foundation for subsequent tasks such as pipeline positioning, tracking, defect detection, and maintenance. However, this detection is often hindered by severe occlusion and extreme lighting conditions. Inspired by human visual reasoning, the detection effect of pipeline edges under challenging scenarios primarily relies on contextual information. Therefore, we propose a context-enhanced deep learning network based on keypoint representation for underwater pipeline edge detection (CEPDNet). The network can effectively capture contextual information by fusing a crisscross attention mechanism (CCAM) that is highly sensitive to the crisscross distribution of underwater pipelines. Additionally, a slope output branch (SOB) is added to the CEPDNet to correlate each keypoint with its adjacent points for local information supplement. To fill the gap of diversified, multiscene, and multiview underwater pipeline edge detection datasets, an annotated dataset (UPEdge) comprising real, generated, and simulated data has been constructed. Notably, an automatic annotation method in a simulated environment is proposed. Experimental results show that our model achieves a detection accuracy of 91.32% on the self-constructed dataset. Moreover, our model demonstrates high deployment friendliness, with a detection speed of 4.9 ms per image postdeployment. In generalization tests, the CEPDNet also achieves competitive performance on the TuSimple dataset.