Title
Multi-Objective Performance Evaluation of the Detection of Catenary Support Components Using DCNNs
Author
Liu, W. (Southwest Jiaotong University)
Liu, Zhigang (Southwest Jiaotong University)
Nunez, Alfredo (TU Delft Railway Engineering) 
Wang, Liyou (Southwest Jiaotong University)
Liu, Kai (Southwest Jiaotong University)
Lyu, Yang (Southwest Jiaotong University)
Wang, H. (Southwest Jiaotong University)
Contributor
De Schutter, Bart (editor)
Ferrara, Antonella (editor)
Date
2018
Abstract
The goal of this paper is to evaluate from a multi-objective perspective the performance on the detection of catenary support components when using state-of-the-art deep convolutional neural networks (DCNNs). The detection of components is the first step towards a complete automatized monitoring system that will provide actual information about defects in the catenary support devices. A series of experiments in an unified test environment for detection of components are performed using Faster-CNN, R-FCN, SSD, and YOLOv2. Through the comparison of different assessment indicators, such as precision, recall, average precision and mean average precision, the detection performance of the different DCNNs methods for the components of the catenary support devices is analyzed, discussed and evaluated. The experiment results show that among all considered methods, R-FCN is the more suitable for the detection of catenary support components.
Subject
Catenary
Railway Systems
Multi-Objective Performance Evaluation
Deep convolutional neural networks (DCNNs
To reference this document use:
http://resolver.tudelft.nl/uuid:d6a81a6c-ce0d-4f37-a7d6-c61f9b22a0f4
DOI
https://doi.org/10.1016/j.ifacol.2018.07.017
Source
Proceedings of 15th IFAC Symposium on Control in Transportation Systems (CTS 2018): Savona, Italy, June 6-8, 2018, 51 (9)
Event
15th IFAC Symposium on Control in Transportation Systems, 2018-06-06 → 2018-06-08, Savona, Italy
Series
IFAC-PapersOnLine, 2405-8963, 51 (9)
Part of collection
Institutional Repository
Document type
conference paper
Rights
© 2018 W. Liu, Zhigang Liu, Alfredo Nunez, Liyou Wang, Kai Liu, Yang Lyu, H. Wang