Hybrid semantic segmentation for tunnel lining cracks based on Swin Transformer and convolutional neural network

Abstract

In the field of tunnel lining crack identification, the semantic segmentation algorithms based on convolution neural network (CNN) are extensively used. Owing to the inherent locality of CNN, these algorithms cannot make full use of context semantic information, resulting in difficulty in capturing the global features of crack. Transformer‐based networks can capture global semantic information, but this method also has the deficiencies of strong data dependence and easy loss of local features. In this paper, a hybrid semantic segmentation algorithm for tunnel lining crack, named SCDeepLab, is proposed by fusing Swin Transformer and CNN in the encoding and decoding framework of DeepLabv3+ to address the above issues. In SCDeepLab, a joint backbone network is introduced with CNN‐based Inverse Residual Block and Swin Transformer Block. The former is used to extract the local detailed information of the crack to generate the shallow feature layer, whereas the latter is used to extract the global semantic information to obtain the deep feature layer. In addition, Efficient Channel Attention enhanced Feature Fusion Module is proposed to fuse the shallow and deep features to combine the advantages of the two types of features. Furthermore, the strategy of transfer learning is adopted to solve the data dependency of Swin Transformer. The results show that the mean intersection over union (mIoU) and mean pixel accuracy (mPA) of SCDeepLab on the data sets constructed in this paper are 77.41% and 84.42%, respectively, which have higher segmentation accuracy than previous CNN‐based and transformer‐based semantic segmentation algorithms.

Pavement Crack Classification and Recognition Algorithm Combined With Tensor Voting and RANSAC

“…𝑠 = 𝛼𝑙 sin𝛼 ⁄ and 𝑘 = 2sin𝛼 𝑙 ⁄ , where 𝛼 is the angle between the line OP and the positive direction of the x-axis. c is the parameter controlling the rate of decay, determined by 𝛿, 𝑐 = −16log[0.1(𝛿 − 1)] π 2 ⁄ . The voting field intensity S of point O at any given point P is determined by (5).…”

“…In acquired road images, due to contrasting grayscale values between cracks and surrounding regions, digital image processing techniques can be employed by setting grayscale thresholds to achieve crack pixel detection and segmentation. Compared with the semantic segmentation network in deep learning algorithms [1][2][3], the threshold method exhibits lower segmentation accuracy and lacks continuity, but segments smaller cracks and saves network training time [4].…”

Satellite Video Remote Sensing for Flood Model Validation

“…He et al., 2023; Q. He et al., 2023; Z. Zhou et al., 2023). Although the adoption of these architectures is still evolving, their robustness in understanding complex scenes which exhibit large variations within the same class and subtle differences between different classes, such as in video remote sensing, is yet to be explored.…”

A method of concrete damage detection and localization based on weakly supervised learning

“…Siriborvornratanakul (2023) developed a novel method known as ThinCrack U-Net by optimizing sampling layers and atrous convolution, providing more accurate results for pixel-level thin crack detection on road surfaces. Zhou et al (2023) proposed a hybrid semantic segmentation algorithm for tunnel lining cracks by integrating Swin Transformer and CNN within the encoder-decoder framework of DeepLabv3+. This offers higher precision than previous semantic segmentation algorithms based on CNN and transformers.…”