Real-Time Rail Safety: A Deep Convolutional Neural Network Approach for Obstacle Detection on Tracks

Identifying obstacles on train tracks in real-time can be challenging due to various factors such as visibility, environmental conditions, and the speed of the train. Accurate and efficient detection of obstacles is essential for ensuring the safety of passengers and avoiding derailments. The object...

Full description

Saved in:

Bibliographic Details
Published in:	2023 4th International Conference on Signal Processing and Communication (ICSPC) pp. 101 - 105
Main Authors:	Jenefa, A, Ande, Aaron, Mounikuttan, Thejas, Anuj, M.D., Jenulin Makros, G, Rejoice, G Rachel, Shalini, T Mary
Format:	Conference Proceeding
Language:	English
Published:	IEEE 23-03-2023
Subjects:	Deep Convolutional Neural Network (DCNN) Object Detection Rail Safety Rails Railway Track Real-time systems Rocks Sensitivity and specificity Signal processing Support vector machines Training data
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Identifying obstacles on train tracks in real-time can be challenging due to various factors such as visibility, environmental conditions, and the speed of the train. Accurate and efficient detection of obstacles is essential for ensuring the safety of passengers and avoiding derailments. The objective of this study was to develop a deep convolutional neural network (DCNN) to identify obstacles in train tracks. In recent years, the use of DCNNs has been widespread for image recognition and classification tasks, however, there has been limited research in the field of obstacle identification in train tracks. The DCNN was trained on a dataset of train track images with and without obstacles. The dataset consisted of over 1000 images with various types of obstacles, including rocks, trees, and other debris. The DCNN was implemented using the TensorFlow and Keras libraries. The DCNN was able to accurately identify obstacles in train tracks with an overall accuracy of 98%. The model also showed high sensitivity and specificity in detecting obstacles in train tracks. This study demonstrates the effectiveness of using DCNNs for obstacle identification in train tracks. The high accuracy of the DCNN shows its potential for practical application in real-world scenarios to improve railway safety. Further research is required to test the performance of the DCNN in different environmental conditionsand to optimize its architecture for better performance.
DOI:	10.1109/ICSPC57692.2023.10125284