Generative Approach for Detecting Small Intrusive Foreign Objects in High-Speed Railway Scenario

Foreign object intrusion into high-speed railway (HSR) catenary systems poses severe operational hazards, making effective detection crucial for safety. Precise detection of these small intrusive objects is essential. However, the lack of datasets and research on foreign object intrusion in HSR scenario brings two major challenges: limited data and low accuracy for detecting small intrusive objects. To address these challenges, this paper introduces a novel generative method for detecting foreign object intrusion. To address data limitations, we use low-rank adaptation to fine-tune a diffusion model, developing a generation-extraction-integration framework that generates true-to-reality HSR images of small intrusive target objects. Furthermore, to enhance the detection of small objects in HSR scenario, we propose a new detection model called SA-YOLO. Based on the YOLOv9 architecture, this model optimizes the backbone network using the star operation, an element-wise multiplication method, and introduces the A-DyS module to improve upsampling through dynamic sampling and attention mechanism. Extensive experiments demonstrate that in the HSR scenario our method outperforms existing state-of-the-art approaches in terms of both generation quality and detection performance, while also showing high robustness.

• Publication year

  2026

• Venue

  IEEE transactions on intelligent transportation systems (Print)

• Publication date

  2026-01-01

• Fields of study

  Computer Science, Engineering

• Identifiers
  DOI 10.1109/TITS.2025.3625181
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Foreign Object Detection Method for Railway Catenary Based on a Scarce Image Generation Model and Lightweight Perception Architecture

  2026cited by this paper
• Defect Detection and Classification of Railway Track System for In-Service MRT in Tropical Regions Using a Contactless TBMS and Adaptive-DBSCAN

  2025cited by this paper
• Spectrum-Enhanced Graph Attention Network for Garment Mesh Deformation

  2025cited by this paper
• Attribution explanations for decision-making in deep lane-change models

  2025cited by this paper
• Multivehicle Cooperative Localization using a TOA-Based Simulated Annealing Extended Kalman Filter in Urban Canyons

  2025cited by this paper
• Multi-Scale CNN-Transformer Hybrid Network for Rail Fastener Defect Detection

  2025cited by this paper
• Boundary Consensus of Networked Hyperbolic Systems of Conservation Laws

  2025cited by this paper
• YOLOv12: Attention-Centric Real-Time Object Detectors

  2025cited by this paper
• Simple Drop-in LoRA Conditioning on Attention Layers Will Improve Your Diffusion Model

  2024cited by this paper
• MobileNetV4 - Universal Models for the Mobile Ecosystem

  2024cited by this paper
• Rewrite the Stars

  2024cited by this paper
• Yolo-3DMM for Simultaneous Multiple Object Detection and Tracking in Traffic Scenarios

  2024cited by this paper
• Improved YOLOv5 Based on Attention Mechanism and FasterNet for Foreign Object Detection on Railway and Airway tracks

  2024cited by this paper
• YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information

  2024influential reference
• Towards Goal-oriented Prompt Engineering for Large Language Models: A Survey

  2024influential reference
• Editable Scene Simulation for Autonomous Driving via Collaborative LLM-Agents

  2024cited by this paper
• WindTrans: Transformer-Based Wind Speed Forecasting Method for High-Speed Railway

  2024cited by this paper
• YOLOv11: An Overview of the Key Architectural Enhancements

  2024cited by this paper
• Diffusion Models for Intelligent Transportation Systems: A Survey

  2024cited by this paper
• Embedded Real-Time Vehicle and Pedestrian Detection Using a Compressed Tiny YOLO v3 Architecture

  2024cited by this paper
• Research on the Method of Foreign Object Detection for Railway Tracks Based on Deep Learning

  2024cited by this paper
• YOLOv8: A Novel Object Detection Algorithm with Enhanced Performance and Robustness

  2024cited by this paper
• YOLOv10: Real-Time End-to-End Object Detection

  2024cited by this paper
• DriveGPT4: Interpretable End-to-End Autonomous Driving Via Large Language Model

  2023cited by this paper
• A Deep Generative Approach for Rail Foreign Object Detections via Semisupervised Learning

  2023cited by this paper
• GLIGEN: Open-Set Grounded Text-to-Image Generation

  2023cited by this paper
• Multifeature Fusion-Based Object Detection for Intelligent Transportation Systems

  2023cited by this paper
• Adding Conditional Control to Text-to-Image Diffusion Models

  2023cited by this paper
• Symmetry-Driven Unsupervised Abnormal Object Detection for Railway Inspection

  2023cited by this paper
• A Comprehensive Survey of AI-Generated Content (AIGC): A History of Generative AI from GAN to ChatGPT

  2023cited by this paper
• GPT-4 Technical Report

  2023influential reference
• LayoutDiffusion: Controllable Diffusion Model for Layout-to-Image Generation

  2023cited by this paper
• MasaCtrl: Tuning-Free Mutual Self-Attention Control for Consistent Image Synthesis and Editing

  2023cited by this paper
• DETRs Beat YOLOs on Real-time Object Detection

  2023cited by this paper
• The Big Data Myth: Using Diffusion Models for Dataset Generation to Train Deep Detection Models

  2023cited by this paper
• SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis

  2023influential reference
• YOLOv5s-D: A Railway Catenary Dropper State Identification and Small Defect Detection Model

  2023cited by this paper
• Learning to Upsample by Learning to Sample

  2023cited by this paper
• Q-Bench: A Benchmark for General-Purpose Foundation Models on Low-level Vision

  2023cited by this paper
• YOLO-CSM-Based Component Defect and Foreign Object Detection in Overhead Transmission Lines

  2023cited by this paper
• Intrusion Detection of Foreign Objects in Overhead Power System for Preventive Maintenance in High-Speed Railway Catenary Inspection

  2022cited by this paper
• YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors

  2022cited by this paper
• Foreign Object Detection for Railway Ballastless Trackbeds: A Semisupervised Learning Method

  2022cited by this paper
• CF-YOLO: Cross Fusion YOLO for Object Detection in Adverse Weather With a High-Quality Real Snow Dataset

  2022cited by this paper
• Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding

  2022cited by this paper
• A Stable Lightweight and Adaptive Feature Enhanced Convolution Neural Network for Efficient Railway Transit Object Detection

  2022cited by this paper
• Imagic: Text-Based Real Image Editing with Diffusion Models

  2022cited by this paper
• Image Segmentation Using Text and Image Prompts

  2021cited by this paper
• Learning Transferable Visual Models From Natural Language Supervision

  2021cited by this paper
• Application of Lightweight Railway Transit Object Detector

  2021cited by this paper
• LoRA: Low-Rank Adaptation of Large Language Models

  2021cited by this paper
• A Review of Yolo Algorithm Developments

  2021cited by this paper
• On the Limitations of Multimodal VAEs

  2021cited by this paper
• Generative Adversarial Network (GAN): a general review on different variants of GAN and applications

  2021cited by this paper
• ControlVAE: Controllable Variational Autoencoder

  2020cited by this paper
• Detection Approach Based on an Improved Faster RCNN for Brace Sleeve Screws in High-Speed Railways

  2020cited by this paper
• S3VAE: Self-Supervised Sequential VAE for Representation Disentanglement and Data Generation

  2020cited by this paper
• Encoding in Style: a StyleGAN Encoder for Image-to-Image Translation

  2020cited by this paper
• Generating Diverse High-Fidelity Images with VQ-VAE-2

  2019cited by this paper
• GENERATIVE ADVERSARIAL NETS

  2018cited by this paper
• A Style-Based Generator Architecture for Generative Adversarial Networks

  2018cited by this paper
• StoryGAN: A Sequential Conditional GAN for Story Visualization

  2018cited by this paper
• Attention is All you Need

  2017cited by this paper
• beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework

  2016cited by this paper
• StackGAN: Text to Photo-Realistic Image Synthesis with Stacked Generative Adversarial Networks

  2016cited by this paper
• U-Net: Convolutional Networks for Biomedical Image Segmentation

  2015cited by this paper
• Traffic Scene-Informed Attribution of Autonomous Driving Decisions

  year unknowncited by this paper

• No citing papers are available for this paper.