Automated vehicle for railway track fault detection
01 Nov 2017-Vol. 263, Iss: 5, pp 052045
TL;DR: In this paper, an automated system for railway track inspection is proposed. But in such type of manual inspection, there are many drawbacks that may result in the poor inspection of the track, due to which accidents may cause in future.
Abstract: For the safety reasons, railroad tracks need to be inspected on a regular basis for detecting physical defects or design non compliances. Such track defects and non compliances, if not detected in a certain interval of time, may eventually lead to severe consequences such as train derailments. Inspection must happen twice weekly by a human inspector to maintain safety standards as there are hundreds and thousands of miles of railroad track. But in such type of manual inspection, there are many drawbacks that may result in the poor inspection of the track, due to which accidents may cause in future. So to avoid such errors and severe accidents, this automated system is designed.Such a concept would surely introduce automation in the field of inspection process of railway track and can help to avoid mishaps and severe accidents due to faults in the track.
Citations
More filters
TL;DR: In this paper , three object detection models: YOLOv5, Faster RCNN, and EfficientDet are compared by testing a dataset of 31 images that contain three different railway track elements (clip, rail, and fishplate), both faulty and non-faulty.
References
More filters
TL;DR: An overview of rail defects and their consequences from the earliest days of railways to the present day can be found in this paper, where the authors present an overview of the rail defects in the early days of railway systems.
Abstract: For about 150 years, the steel rail has been at the very heart of the world's railway systems. The rail works in a harsh environment and, as part of the track structure, it has little redundancy; thus, its failure may lead to catastrophic derailment of vehicles, the consequences of which can include death, injury, costs and loss of public confidence. These can have devastating and long-lasting effects on the industry. Despite the advances being made in railway permanent way engineering, inspection and rail-making technology, continually increasing service demands have resulted in rail failure continuing to be a substantial economic burden and a threat to the safe operation of virtually every railway in the world. This paper presents an overview of rail defects and their consequences from the earliest days of railways to the present day.
372 citations
TL;DR: This paper investigates the use of a low-cost camera vision solution capable of urban, rural, or off-road classification based on the analysis of color and texture features extracted from a driver's perspective camera view to resolve two road-type classification problems of varying difficulty.
Abstract: The ongoing development autonomous vehicles and adaptive vehicle dynamics present in many modern vehicles has generated a need for road environment classification - i.e., the ability to determine the nature of the current road or terrain environment from an onboard vehicle sensor. In this paper, we investigate the use of a low-cost camera vision solution capable of urban, rural, or off-road classification based on the analysis of color and texture features extracted from a driver's perspective camera view. A feature set based on color and texture distributions is extracted from multiple regions of interest in this forward-facing camera view and combined with a trained classifier approach to resolve two road-type classification problems of varying difficulty - {off-road, on-road} environment determination and the additional multiclass road environment problem of {off-road, urban, major/trunk road and multilane motorway/carriageway}. Two illustrative classification approaches are investigated, and the results are reported over a series of real environment data. An optimal performance of ~90% correct classification is achieved for the {off-road, on-road} problem at a near real-time classification rate of 1 Hz.
81 citations