S
Sung-Jin Song
Researcher at Sungkyunkwan University
Publications - 196
Citations - 1607
Sung-Jin Song is an academic researcher from Sungkyunkwan University. The author has contributed to research in topics: Ultrasonic sensor & Ultrasonic testing. The author has an hindex of 17, co-authored 188 publications receiving 1335 citations. Previous affiliations of Sung-Jin Song include Iowa State University & Korea University.
Papers
More filters
Book
Ultrasonic Nondestructive Evaluation Systems: Models and Measurements
Lester W. Schmerr,Sung-Jin Song +1 more
TL;DR: In this article, the authors present a comprehensive model of an ultrasonic measurement system, including Fourier analysis, linear system theory, and wave propagation and scattering theory, using MATLAB examples and exercises.
Journal ArticleDOI
Convolutional neural network for ultrasonic weldment flaw classification in noisy conditions
TL;DR: The convolutional neural network is applied to noisy ultrasonic signatures to improve classification performance of weldment defects and applicability and the result shows that CNN is robust, does not require specific feature extraction methods and give considerable high defect classification accuracies even for noisy signals.
Journal ArticleDOI
Development of an ultra sonic phased array system for nondestructive tests of nuclear power plant components
TL;DR: A phased array ultrasonic inspection (PAULI) system was developed to obtain electronically scanned ultrasonic images of the inside of nuclear power plant components for nondestructive evaluation.
Journal ArticleDOI
Performance enhancement of convolutional neural network for ultrasonic flaw classification by adopting autoencoder
TL;DR: The results demonstrate that the autoencoder can successfully remove noise from the ultrasonic weldment defect signals, which consequently improve the defect classification accuracy of the artificially intelligent deep learning classifiers.
Journal ArticleDOI
Investigation of deep neural network with drop out for ultrasonic flaw classification in weldments
TL;DR: Data driven approaches are used to train the neural network for defect classification without extracting any feature from ultrasonic signals and the results demonstrate that given deep neural network architecture is more robust and the network can classify defects with high accuracy without extractingAny feature from Ultrasonic signals.