Automatic Rail Component Detection Based on AttnConv-Net

TL;DR

This paper introduces AttnConv-Net, an attention-based deep learning model that improves the detection of rail components in images, offering faster and more accurate results without complex pre- or post-processing.

Contribution

The paper presents a novel attention-powered deep convolutional network with cascading attention blocks for rail component detection, simplifying the detection pipeline and enhancing small object detection.

Findings

Outperforms classic CNN methods in accuracy and speed

Effectively detects multiple rail components including small parts

Validated on real and synthetic datasets

Abstract

The automatic detection of major rail components using railway images is beneficial to ensure the rail transport safety. In this paper, we propose an attention-powered deep convolutional network (AttnConv-net) to detect multiple rail components including the rail, clips, and bolts. The proposed method consists of a deep convolutional neural network (DCNN) as the backbone, cascading attention blocks (CAB), and two feed forward networks (FFN). Two types of positional embedding are applied to enrich information in latent features extracted from the backbone. Based on processed latent features, the CAB aims to learn the local context of rail components including their categories and component boundaries. Final categories and bounding boxes are generated via two FFN implemented in parallel. To enhance the detection of small components, various data augmentation methods are employed in the training process. The effectiveness of the proposed AttnConv-net is validated with one real dataset and another synthesized dataset. Compared with classic convolutional neural network based methods, our proposed method simplifies the detection pipeline by eliminating the need of prior- and post-processing, which offers a new speed-quality solution to enable faster and more accurate image-based rail component detections