Improved SOLOv2 detection method for shield tunnel lining water leakages

Abstract

With the rapid development of urban metros, the detection of shield tunnel leakages has become an important research topic. Progressive technological innovations such as deep learning-based methods provide an effective way to detect tunnel leakages accurately and automatically. However, due to the complex shapes and sizes of leakages, it is challenging for existing algorithms to detect such defects. To address these issues, this paper proposes a novel deep learning-based model named segmenting objects by locations network v2 for tunnel leakages (SOLOv2-TL), which is enhanced by ResNeXt-50, deformable convolution, and path augmentation feature pyramid network (PAFPN). In the SOLOv2-TL, ResNeXt-50 coupled with deformable convolution is the backbone for boosting feature extraction ability that would enable the model sensitivity to leakages of different shapes. The PAFPN is introduced as the neck to reduce the loss of leakage information and more accurately assign leakages of different sizes to their corresponding feature levels. The superior performances of ResNeXt-50 with deformable convolution and PAFPN were validated by ablation tests. Moreover, the segmentation results obtained by SOLOv2-TL were compared with those by the mask region-based convolutional neural network (Mask R-CNN), Cascade Mask R-CNN, and SOLO which demonstrated that the mAP, mAP 50 , and mAP 75 of SOLOv2-TL are higher than those of the other methods, where mAP indicates the mean mask average precision (AP) at intersection over union (IoU) = 0.50:0.05:0.95, mAP 50 refers to the mean mask AP with an IoU threshold of 0.50, and mAP 75 denotes the mean mask AP with an IoU threshold of 0.75. Finally, a leakage area quantification method is presented.

Effect of groundwater decline on plant induced by tunnel excavation and calculation of ecological water level based on SPAC model

“…High underground water pressure and corrosive ions in groundwater can affect the safety and durability of tunnels [1][2][3]. In the past tunnel construction, the treatment of groundwater was mainly based on drainage, which could significantly reduce the water pressure of the tunnel lining [4], prevent water inrush [5], and ensure the safety of the tunnel structure [6]. However, when the groundwater was drained for a long time and was not replenished in time, the groundwater level in the tunnel site would decline.…”

Intelligent recognition of voids behind tunnel linings using deep learning and percussion sound

“…To cope with this problem, the public has shifted more attention to the development and utilization of underground space [1]. Traffic tunnels, as important infrastructures within the underground space, play a crucial role in relieving urban congestion and are being extensively constructed in many countries [2,3]. As more and more traffic tunnels are put into service, under the combined influence of external disturbances and the deterioration of the tunnel structures, structural defects, such as water leakages [4,5], cracks [6], and voids [7], will occur in tunnels during the operation and maintenance period.…”