eduzhai > Applied Sciences > Engineering >

Machine learning in problems of automation of ultrasound diagnostics of railway tracks

  • Save

... pages left unread,continue reading

Document pages: 7 pages

Abstract: The article presents the system architecture for automatic decoding ofrailway track defectograms in real time. The system includes an ultrasound datapreprocessing module, a set of neutral network classifiers, a decision block.Preprocessing of data includes affine transformations of measurementinformation into a format suitable for the operation of a neural network, aswell as a combination of information on measurement channels, depending on thetype of defect being defined. The classifier is built on a convolutional neuralnetwork. The proposed solution can be effectively implemented on a modernelemental basis for performing parallel computing, including tensor processorand GPUs.

Please select stars to rate!


0 comments Sign in to leave a comment.

    Data loading, please wait...