Railway defect detection based on track geometry using supervised and unsupervised machine learning

Influences of Railway Track Geometry Parameters on Ground-Borne Vibration

One of the most important challenges in the development of urban railways network is the effects of railway vibrations on human and adjacent structures. Although various investigations into the sources of railway vibrations have been conducted, very few research works have been reported on the influences of track geometry parameters on the level of vibrations. This is addressed in this resea...

Novel strategies on semi-supervised and unsupervised machine learning

Using machine learning for defect detection

In this paper we present an approach to defect detection in images based on machine learning algorithms. A qualitative model of defect has been devised based on human experience. A set of vision primitives measuring defect features has been deened. Objects in images candidate to be classiied as defects are submitted to automatic classiication, which is achieved with learning by examples algorit...

Influences of Track Structure, Geometry and Traffic Parameters on Railway Deterioration

The roles of the several parameters that influences railway track deterioration most, are examined in this research with a view to make railway track maintenance more effective and cost efficient. The results presented are based on a comprehensive study of railway track degradation on super structure, sub-structure and geometrical aspects. The changes in TQI (Track Quality Index), the track set...

Vision Based Railway Track Monitoring using Deep Learning

Computer vision based methods have been explored in the past for detection of railway track defects, but full automation has always been a challenge because both traditional image processing methods and deep learning classifiers trained from scratch fail to generalize that well to infinite novel scenarios seen in the real world, given limited amount of labeled data. Advancements have been made ...