Title: Novel freight train image fault detection and classification models based on CNN

Authors: Longxin Zhang; Yang Hu; Tianyu Chen; Hong Wen; Peng Zhou; Wenliang Zeng

Addresses: College of Computer Science, Hunan University of Technology, Zhuzhou, Hunan, China ' College of Computer Science, Hunan University of Technology, Zhuzhou, Hunan, China ' College of Computer Science, Hunan University of Technology, Zhuzhou, Hunan, China ' College of Computer Science, Hunan University of Technology, Zhuzhou, Hunan, China ' College of Computer Science, Hunan University of Technology, Zhuzhou, Hunan, China ' College of Computer Science, Hunan University of Technology, Zhuzhou, Hunan, China

Abstract: The existing freight train detection model could not meet the demand of actual applications. Aiming at the problem of typical train image fault detection of freight trains, a multi-class freight train (MFT) fault recognition model is proposed in this study. First, an object detection model is designed to reduce the dependence on colour and texture, and a bounding box regression method is used to select candidate boxes. Second, a fault classification model is developed to classify the segmented image. Experimental results on real train images show that the mean accuracy rate (mAR) of MFT for typical faults can reach 92.55%, which is 9.83% and 4.62% higher than those of the traditional machine learning and state-of-the-art deep learning methods, and has good anti-interference ability for image rotation and noise. In addition, the mAR of MFT on the public dataset can reach 94.60%, and it also has good recognition performance.

Keywords: convolutional neural network; deep learning; fault detection; freight train fault; image classification.

DOI: 10.1504/IJCSE.2023.133690

International Journal of Computational Science and Engineering, 2023 Vol.26 No.5, pp.567 - 578

Received: 29 May 2022
Received in revised form: 05 Oct 2022
Accepted: 28 Oct 2022
Published online: 29 Sep 2023 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article