A method for automatic detection and identification of ballasted track defects

Abstract

The invention provides an automatic detection and identification method for ballasted track defects, relates to the technical field of railway defect detection, and designs a sliding time window with a window size of 512×512 and a window step size of 512trace. The hyperbolic asymptote direction features in the time window are extracted by using the energy feature extraction method of the statistical domain, and whether there is a disease in the window is judged, identified and recorded according to the Euclidean distance space. Move the sliding window, judge, identify and record whether there is a disease in the next window, and cycle in turn to realize the automatic detection and identification of ballasted track diseases. The method realizes an automatic detection method for ballasted track defects, greatly reduces labor costs, improves detection efficiency, and has strong practicability.

Description

technical field

[0001] The invention relates to the technical field of railway disease detection, in particular to an automatic detection and identification method for ballasted track diseases. Background technique

[0002] Diseases such as slag pockets and mud slopping in the middle of the ballasted track have seriously affected the safety and comfort of driving. The current detection methods mainly include elastic wave method and full wave field method. These methods have slow detection speed and high labor cost, and are not suitable for large-scale general inspection. As a rapid and continuous non-destructive testing technology, ground penetrating radar is an important means to detect ballasted track defects. However, the existing GPR echo signal interpretation method is mainly through manual methods. Due to the different experience of each person, it is easy to cause missed or false detection of diseases. Contents of the invention

[0003] An embodiment of the prese...

Images