Detection Tasks

For more details about definition of these tasks, please refer to:

    Gao, Y. and Mosalam, K. M. (2018), Deep Transfer Learning for Image‐Based Structural Damage Recognition. Computer-Aided Civil and Infrastructure Engineering. 2018. doi:10.1111/mice.12363 [link]


    1. Scene level: 3 classes (pixel/object/structural levels)

    Pixel level Object level Structural level


    2. Damage state: 2 classes (undamaged/damaged)

    Undamaged Damaged


    3. Spalling condition: 2 classes (no spalling/spalling)

    Non-spalling Spalling


    4. Material type: 2 classes(steel/others)

    4_a 4_b
    Steel Others


    5. Collapse mode: 3 classes (no/partial collapse/collapse)

    5_a 5_b 5_c
    Non-collapse Partial collapse Global collapse


    6. Component type: 4 classes (beam/column/wall/else)

    6_a 6_b 6_c 6_d
    Beam Column Wall Else (i.e.joint)


    7. Damage level: 4 classes (no/minor/moderate/heavy damage)

    7_a 7_b 7_c 7_d
    No damage Minor damage Moderate damage Heavy damage


    8. Damage type: 4 classes (no/flexural/shear/combined damage)

    8_a 8_b 8_c 8_d
    No damage Flexural damage Shear damage Combined damage