Non-destructive testing (NDT) is the process of analyzing the materials, components, structures, etc. without causing damage to it. In this paper, a NDT technique is proposed for detecting the cracks in the concrete surfaces. Initial results are obtained using a neural network for concrete crack detection. A convolution neural network model has been developed and trained using both positive (crack images) and negative (non-crack) images. In this work, a database consisting of 40,000 images is used. The model is trained with 36,000 images, 4000 for validation and 4000 for testing. To evaluate the effectiveness of model, accuracy, recall, precision and F1 score parameters are calculated.
CITATION STYLE
Koppad, D., & Paramanandham, N. (2020). Non-destructive testing for cracks in concrete. In Lecture Notes in Electrical Engineering (Vol. 656, pp. 657–664). Springer. https://doi.org/10.1007/978-981-15-3992-3_56
Mendeley helps you to discover research relevant for your work.