LPCAN: Lightweight Pyramid Cross-Attention Network for Rail Surface Defect Detection Using RGB-D Data

TL;DR

LPCANet is a lightweight, efficient neural network that combines RGB-D data and novel modules to accurately detect rail surface defects, outperforming existing methods in speed and accuracy.

Contribution

The paper introduces LPCANet, a novel lightweight pyramid cross-attention network that effectively fuses RGB-D data for rail defect detection with superior performance and efficiency.

Findings

Achieves state-of-the-art accuracy with only 9.90 million parameters.

Demonstrates high inference speed of 162.60 fps.

Outperforms 18 existing methods on multiple datasets.

Abstract

This paper addresses the limitations of current vision-based rail defect detection methods, including high computational complexity, excessive parameter counts, and suboptimal accuracy. We propose a Lightweight Pyramid Cross-Attention Network (LPCANet) that leverages RGB-D data for efficient and accurate defect identification. The architecture integrates MobileNetv2 as a backbone for RGB feature extraction with a lightweight pyramid module (LPM) for depth processing, coupled with a cross-attention mechanism (CAM) for multimodal fusion and a spatial feature extractor (SFE) for enhanced structural analysis. Comprehensive evaluations on three unsupervised RGB-D rail datasets (NEU-RSDDS-AUG, RSDD-TYPE1, RSDD-TYPE2) demonstrate that LPCANet achieves state-of-the-art performance with only 9.90 million parameters, 2.50 G FLOPs, and 162.60 fps inference speed. Compared to 18 existing methods, LPCANet shows significant improvements, including +1.48\% in $S_{\alpha}$, +0.86\% in IOU, and +1.77\% in MAE over the best-performing baseline. Ablation studies confirm the critical roles of CAM and SFE, while experiments on non-rail datasets (DAGM2007, MT, Kolektor-SDD2) validate its generalization capability. The proposed framework effectively bridges traditional and deep learning approaches, offering substantial practical value for industrial defect inspection. Future work will focus on further model compression for real-time deployment.