Deep Multitask Learning for Railway Track Inspection

By Xavier Gibert, Vishal M. Patel, Ramalingam Chellappa. IEEE transactions on intelligent transportation systems (Print), 2015

393 citations recorded by the source corpus.

Railroad tracks need to be periodically inspected and monitored to ensure safe transportation. Automated track inspection using computer vision and pattern recognition methods has recently shown the potential to improve safety by allowing for more frequent inspections while reducing human errors. Achieving full automation is still very challenging due to the number of different possible failure modes, as well as the broad range of image variations that can potentially trigger false alarms. In addition, the number of defective components is very small, so not many training examples are available for the machine to learn a robust anomaly detector. In this paper, we show that detection performance can be improved by combining multiple detectors within a multitask learning framework. We show that this approach results in improved accuracy for detecting defects on railway ties and fasteners.

Subfields: Civil and Structural Engineering, Mechanical Engineering, Media Technology

View Paper: https://www.semanticscholar.org/paper/371322935e635ff56c10a74395b6b66a7487fa83

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Civil and Structural Engineering

In Mechanical Engineering

In Media Technology

Citations in the corpus

C2AE: Class Conditioned Auto-Encoder for Open-Set Recognition (2019, methodology): “It has been used in many applications such as ﬁnance, railway track inspection etc. [23], [1], [9] as well as open-set recognition [4], [36], [40].”
Automatic Metallic Surface Defect Detection and Recognition with Convolutional Neural Networks (2018, background): “These machine vision-based inspection systems occur in many industrial applications, such as steel strip inspection [2,3], liquid crystal display (LCD) inspection [4], fabric inspection [5,6], aluminum profiles [7], railway track inspection [8], food inspection [9], and optical components inspection”
Deep Learning for Smart Industry: Efficient Manufacture Inspection System With Fog Computing (2018, methodology): “[13] adopt multiple detectors to improve accuracy for detecting defects on components of the railway.”
Deep convolutional neural networks for detection of rail surface defects (2016, methodology): “While [21], [22] have focused on identification of track components such as ballast, concrete, wood, and fastener, in this paper we focus on the detection and classification of different defects that occur at the rail surface.”
Deep Metallic Surface Defect Detection: The New Benchmark and Detection Network (2020, background): “There are lots of other features, such as co-occurrence matrix (GLCM) [23] and some grayscale statistical features [24,25].”
Potential, challenges and future directions for deep learning in prognostics and health management applications (2020, methodology): “Gibert et al. (2015) built a multi-task deep CNN using image data collected for the purpose of monitoring railway infrastructure.”
Automatic Defect Detection of Fasteners on the Catenary Support Device Using Deep Convolutional Neural Network (2018, background): “As deep convolutional neural network (DCNN) [11] prevails in object recognition, Gibert et al. [12] applied DCNN in railroad track detection.”
Emerging materials intelligence ecosystems propelled by machine learning (2020, background): “In more industrial settings, automated image analysis has also led to easy identification of defects, cracks or corrosion in structural materials, such as railway tracks and asphalt pavements 167 .”