Automatic sewer defect detection and severity quantification based on pixel-level semantic segmentation

Construction and maintenance of urban underground infrastructure with digital technologies

“…Compared with traditional workflow, deep learning-based methods are more straightforward by training the model with annotated data and then applying the trained model to generate results. CNN-based models are leveraged to facilitate different levels of visual interpretation tasks for underground infrastructure, including defect classification [58][59][60], detection [45,[61][62][63][64], and segmentation [65][66][67][68]. Among them, classification is an image-level task to recognize if the image contains defects or predict the defect type for the whole image, while detection is object-level, aiming to simultaneously classify defect type and localize location with bounding boxes for each defect.…”

Toward Field Soil Surveys: Identifying and Delineating Soil Diagnostic Horizons Based on Deep Learning and RGB Image

“…All validation indicators would be performed at the pixel level. In studies related to semantic segmentation, loss functions are mostly used for the convergence of the model, whereas for the final evaluation of the model, mean intersection over union (MIoU) and pixel accuracy (PA) are most widely used [51][52][53].…”

Multi-scale hybrid vision transformer and Sinkhorn tokenizer for sewer defect classification

“…However, the main focus of the field within recent years has been on the defect detection and segmentation tasks [30][31][32]17,[33][34][35][36], where no public datasets are available. The field has, however, become more transparent as many have started to directly compare different methods on the same datasets, in an effort to offset the lack of public detection and segmentation datasets [17,36,34].…”

“…The field has, however, become more transparent as many have started to directly compare different methods on the same datasets, in an effort to offset the lack of public detection and segmentation datasets [17,36,34]. Recently, the field has also started investigating other parts of the sewer inspection process [30,32,17,[37][38][39][40][41], such as Haurum et al [37] proposing a multi-task classification approach for simultaneously classifying defects, water level, pipe material, and pipe shape, and Wang et al [30] proposed a framework to accurately determine the severity of defects related to the operation and maintenance of the pipes. The field has also adopted recent trends from the general computer vision field such as selfsupervised learning [39], synthetic data generation [25,24,[42][43][44], neural architecture search [45], and usage of the Transformer architecture [17,46], indicating that the automated sewer inspection field is catching up to the general computer vision domain.…”