**5. Conclusions**

FS criteria = gini Number of DT = 150

A deep learning CNN model with 11 trainable hidden layers was proposed to automatically extract and classify the bridge damage represented by the continuous deflection of bridge. Although current research on the use of FOG-based test technique to detect the damage of a scale-down bridge model through deep learning is just a pilot study, the following conclusions can be drawn:


**Funding:** This research was funded by the National Natural Science Foundation of China grant number 61875155.

**Author Contributions:** Data curation, X.Z.; Funding acquisition, S.L.; Methodology, S.L.; Project administration, Z.L.; Resources, H.W.; Supervision, H.W.; Writing—original draft, X.Z.; Writing—review & editing, S.L. All authors have read and agreed to the published version of the manuscript

**Acknowledgments:** The research work reported in this paper was supported by the National Engineering Laboratory for Fiber Optic Sensing Technology, Wuhan University of Technology.

**Conflicts of Interest:** The authors declare no conflict of interest.
