Close
Metadata

%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2021/09.06.21.01
%2 sid.inpe.br/sibgrapi/2021/09.06.21.01.28
%T A Vision-based Solution for Track Misalignment Detection
%D 2021
%A Jerripothula, Koteswar Rao,
%A Ansari, Sharik Ali,
%A Nijhawan, Rahul,
%@affiliation Indraprastha Institute of Information Technology Delhi (IIIT-Delhi)
%@affiliation College of Engineering Roorkee (COER)
%@affiliation University of Petroleum and Energy Studies (UPES)
%E Paiva, Afonso,
%E Menotti, David,
%E Baranoski, Gladimir V. G.,
%E Proença, Hugo Pedro,
%E Junior, Antonio Lopes Apolinario,
%E Papa, João Paulo,
%E Pagliosa, Paulo,
%E dos Santos, Thiago Oliveira,
%E e Sá, Asla Medeiros,
%E da Silveira, Thiago Lopes Trugillo,
%E Brazil, Emilio Vital,
%E Ponti, Moacir A.,
%E Fernandes, Leandro A. F.,
%E Avila, Sandra,
%B Conference on Graphics, Patterns and Images, 34 (SIBGRAPI)
%C Gramado (Virtual), Brazil
%8 October 18th to October 22nd, 2021
%I IEEE Computer Society
%J Los Alamitos
%S Proceedings
%K railway, transfer learning, VGG, Inception.
%X Derailment is one of the most frequent ways railway accidents happen. Track defects such as buckling and hogging that cause misalignment of tracks can easily lead to derailments. While railway tracks get laterally misaligned due to buckling, vertical misalignments can result from hogging. Such misalignments are visibly recognizable, and we can even automate recognition using data-driven models. This paper discusses how we build such data-driven models. There are no public datasets available to build such models; therefore, we introduce TMD (Track Misalignment Detection) dataset. It consists of misaligned and normal track images. The problem we try to solve here is essentially a binary image classification problem, which we solve by exploring the feature extraction approach to transfer learning (TL). In this approach, we employ a pre-trained network to extract rich features, which are then supplied with annotations to a learning algorithm for building a candidate TL model. Several pre-trained networks and learning algorithms exist, resulting in multiple candidate TL models; therefore, it becomes essential to identify effective ones. We propose an evaluation criterion to decide which are effective ones using our proposed bias-variance analysis. Our experiments demonstrate that the candidate TL models selected based on our proposed evaluation criterion perform better than other candidate TL models while testing.
%@language en
%3 SIBGRAPI_Railway (3).pdf


Close