Edge-Enabled Real-time Railway Track Segmentation

TL;DR

This paper presents a lightweight, edge-optimized railway track segmentation algorithm that combines Ghost convolution, quantization, and parallel processing to achieve real-time performance on resource-limited devices.

Contribution

It introduces a novel combination of Ghost convolution, a lightweight detection head, quantization, and parallel processing tailored for edge devices in railway track segmentation.

Findings

Achieves 83.3% accuracy on RailSem19 dataset.

Runs at 25 frames per second on Jetson Nano.

Effectively balances accuracy and efficiency for real-time edge applications.

Abstract

Accurate and rapid railway track segmentation can assist automatic train driving and is a key step in early warning to fixed or moving obstacles on the railway track. However, certain existing algorithms tailored for track segmentation often struggle to meet the requirements of real-time and efficiency on resource-constrained edge devices. Considering this challenge, we propose an edge-enabled real-time railway track segmentation algorithm, which is optimized to be suitable for edge applications by optimizing the network structure and quantizing the model after training. Initially, Ghost convolution is introduced to reduce the complexity of the backbone, thereby achieving the extraction of key information of the interested region at a lower cost. To further reduce the model complexity and calculation, a new lightweight detection head is proposed to achieve the best balance between accuracy and efficiency. Subsequently, we introduce quantization techniques to map the model's floating-point weights and activation values into lower bit-width fixed-point representations, reducing computational demands and memory footprint, ultimately accelerating the model's inference. Finally, we draw inspiration from GPU parallel programming principles to expedite the pre-processing and post-processing stages of the algorithm by doing parallel processing. The approach is evaluated with public and challenging dataset RailSem19 and tested on Jetson Nano. Experimental results demonstrate that our enhanced algorithm achieves an accuracy level of 83.3% while achieving a real-time inference rate of 25 frames per second when the input size is 480x480, thereby effectively meeting the requirements for real-time and high-efficiency operation.