Improvement of impact-echo method by applying image recognition of sound spectrogram

H. Shimbo, J.-I. Nojima, T. Mizobuchi (2019). "Improvement of impact-echo method by applying image recognition of sound spectrogram." In Proceedings of FraMCoS 10, Bayonne, France.

H. Shimbo, J.-I. Nojima, T. Mizobuchi. "Improvement of impact-echo method by applying image recognition of sound spectrogram." Paper presented at FraMCoS 10, Bayonne, France, 2019.

H. Shimbo, J.-I. Nojima, T. Mizobuchi, "Improvement of impact-echo method by applying image recognition of sound spectrogram," in Proc. FraMCoS 10, Bayonne, France, 2019.

@inproceedings{h-shimbo2019,
  title     = {Improvement of impact-echo method by applying image recognition of sound spectrogram},
  author    = {H. Shimbo, J.-I. Nojima, T. Mizobuchi},
  year      = {2019},
  booktitle = {Proceedings of FraMCoS 10},
  address   = {Bayonne, France},
}

TY  - CPAPER
TI  - Improvement of impact-echo method by applying image recognition of sound spectrogram
AU  - H. Shimbo
AU  - J.-I. Nojima
AU  - T. Mizobuchi
PY  - 2019
T2  - Proceedings of FraMCoS 10
CY  - Bayonne, France
UR  - https://framcos.org/wp-content/uploads/framcos-papers/Improvement_of_impact-echo_method_by_applying_image_recognition_of_sound_spectro.pdf
AB  - The authors are investigating a method for lifetime prediction of concrete structures by a comprehensive judgment of survey results and information on design and environment. In this paper, we tried to apply impact-echo acoustic spectrogram to image recognition by machine learning in order to have a better estimation of defects in concrete. impact-echo sound sampling was conducted on specimens containing variated voids in depth and diameter. Each sound spectrogram was transformed into 28 by 28 pixels grayscale picture. As a result of the study, it was confirmed that the estimation by machine learning showed competitive accuracy as estimation by a…
ER  -