Semantic segmentation of cracks: Data challenges and architecture

被引:0
|
作者
Panella, Fabio [1 ]
Lipani, Aldo [1 ]
Boehm, Jan [1 ]
机构
[1] Dept. of Civil, Environ. and Geomatic Eng., UCL, London,UK, United Kingdom
基金
英国工程与自然科学研究理事会;
关键词
Deep learning - Crack detection - Network layers - Semantics - Network architecture - Architecture;
D O I
暂无
中图分类号
学科分类号
摘要
Deep Learning (DL) semantic image segmentation is a technique used in several fields of research. The present paper analyses semantic crack segmentation as a case study to review the up to date research on semantic segmentation in the presence of fine structures and the effectiveness of established approaches to address the inherent class imbalance issue. The established UNet architecture is tested against networks consisting exclusively of stacked convolution without pooling layers (straight networks), with regard to the resolution of their segmentation results. Dice and Focal losses are also compared against each other to evaluate their effectiveness on highly imbalanced data. With the same aim, dropout and data augmentation approaches are tested, as additional regularizing mechanisms, to address the uneven distribution of the dataset. The experiments show that the good selection of the loss function has more impact in handling the class imbalance and boosting the detection performance than all the other regularizers with regards to segmentation resolution. Moreover, UNet, the architecture considered as reference, clearly outperforms the networks with no pooling layers both in performance and training time. The authors argue that UNet architectures, compared to the networks with no pooling layers, achieve high detection performance at a very low cost in terms of training time. Therefore, the authors consider such architecture as the state of the art for semantic segmentation of cracks. On the other hand, once computational cost is not an issue anymore thanks to constant improvements of technology, the application of networks without pooling layers might become attractive again because of their simplicity of and high performance. © 2022 Elsevier B.V.
引用
收藏
相关论文
共 50 条
  • [41] Waterfall Atrous Spatial Pooling Architecture for Efficient Semantic Segmentation
    Artacho, Bruno
    Savakis, Andreas
    SENSORS, 2019, 19 (24)
  • [42] Custom CDGNet Architecture for Precise Human Part Semantic Segmentation
    Lovanshi, Mayank
    Tiwari, Vivek
    Ingle, Rajesh
    Jain, Swati
    Lecture Notes in Networks and Systems, 2024, 994 LNNS : 343 - 354
  • [43] Synthetic Data for Semantic Segmentation in Underwater Imagery
    Pergeorelis, Michael
    Bazik, Maxim
    Saponaro, Philip
    Kim, Joong
    Kambhamettu, Chandra
    2022 OCEANS HAMPTON ROADS, 2022,
  • [44] Semantic Segmentation of Fused Mobile Mapping Data
    Villinger, Georg
    Schmitt, Annette
    Reiterer, Alexander
    OPTICAL SENSING AND DETECTION VIII, 2024, 12999
  • [45] Semantic Segmentation of Hippocampal Subregions With U-Net Architecture
    Nasser, Soraya
    Naoui, Moulkheir
    Belalem, Ghalem
    Mahmoudi, Said
    INTERNATIONAL JOURNAL OF E-HEALTH AND MEDICAL COMMUNICATIONS, 2021, 12 (06)
  • [46] HOUGHENCODER: NEURAL NETWORK ARCHITECTURE FOR DOCUMENT IMAGE SEMANTIC SEGMENTATION
    Sheshkus, Alexander
    Nikolaev, Dmitry
    Arlazarov, Vladimir L.
    2020 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP), 2020, : 1946 - 1950
  • [47] Quantized Semantic Segmentation Deep Architecture for Deployment on an Edge Computing Device for Image Segmentation
    Ahamad, Afaroj
    Sun, Chi-Chia
    Kuo, Wen-Kai
    ELECTRONICS, 2022, 11 (21)
  • [48] Buffer ladder feature fusion architecture for semantic segmentation improvement
    Zonghui Liu
    Zhichun Lei
    Signal, Image and Video Processing, 2024, 18 : 475 - 483
  • [49] Circle-U-Net: An Efficient Architecture for Semantic Segmentation
    Sun, Feng
    Kumar, V. Ajith
    Yang, Guanci
    Zhang, Ansi
    Zhang, Yiyun
    ALGORITHMS, 2021, 14 (06)
  • [50] Semantic Segmentation from Limited Training Data
    Milan, A.
    Pham, T.
    Vijay, K.
    Morrison, D.
    Tow, A. W.
    Liu, L.
    Erskine, J.
    Grinover, R.
    Gurman, A.
    Hunn, T.
    Kelly-Boxall, N.
    Lee, D.
    McTaggart, M.
    Rallos, G.
    Razjigaev, A.
    Rowntree, T.
    Shen, T.
    Smith, R.
    Wade-McCue, S.
    Zhuang, Z.
    Lehnert, C.
    Lin, G.
    Reid, I.
    Corke, P.
    Leitner, J.
    2018 IEEE INTERNATIONAL CONFERENCE ON ROBOTICS AND AUTOMATION (ICRA), 2018, : 1908 - 1915