北京交通大学学报2025，Vol.49Issue(3)：56-67,12.DOI:10.11860/j.issn.1673-0291.20230145
基于特征分治与融合的铁路扣件轻量化实时检测模型
Lightweight and real-time detection model for railway fasteners based on feature divide-and-conquer and fusion
摘要
Abstract
To address the challenge of balancing accuracy and detection speed in processing large-scale visual image data of railway fasteners on embedded devices in real time,a lightweight real-time detec-tion model based on attention divide-and-conquer and feature fusion is proposed.First,a hybrid divide-and-conquer attention module,leveraging both spatial and channel features,is introduced to en-hance the model's feature extraction capability and reduce interference from complex backgrounds in the image.Second,a dual divide-and-conquer feature fusion method is designed to improve detection performance across targets of varying sizes.In the construction of the cost volume detection head(YOLO Head),a Varifocal Loss(VFL)function is employed to replace the binary cross-entropy loss used in YOLOX-Nano,thereby enhancing the accuracy of lightweight real-time detection.Further-more,a Random Alpha-IoU(RAL)loss function is adopted to dynamically adjust parameters,slow down convergence,and optimize the training curve,thus preventing the model from falling into local optima.Finally,a dataset of 10,233 annotated fastener targets categorized into six types is used for evaluation.Comparative experiments are conducted using mainstream object detection models,includ-ing YOLOX-Nano,Faster R-CNN,and YOLOv8n.The research results indicate that the proposed model achieves a frame rate of 60.24 Frames Per Second(FPS)and an Average Precision(AP)of 83.40％,representing a 3.24％improvement over the baseline.The parameter count is 2.31 M,which is 54.08％fewer than YOLOX Tiny,and the floating-point operations is 1.99 G,a 69.15％de-crease compared to YOLOX Tiny.These findings provide valuable insights for the development of lightweight real-time detection models and embedded computing systems.关键词
轻量级嵌入式系统/分治混合注意力模块/分治特征融合/代价体构建Key words
lightweight embedded system/divide-and-conquer hybrid attention module/divide-and-conquer feature fusion/cost volume construction分类
信息技术与安全科学引用本文复制引用
鄢化彪,林初欣,黄绿娥,李东丽,刘词波,徐方奇..基于特征分治与融合的铁路扣件轻量化实时检测模型[J].北京交通大学学报,2025,49(3):56-67,12.基金项目
国家自然科学基金(62001202) （62001202）
江西省自然科学基金(20224BAB202036) （20224BAB202036）
江西省教育厅科学技术重点研究项目(GJJ2200805) （GJJ2200805）
江西省研究生创新专项资金(YC2022-S692)National Natural Science Foundation of China(62001202) （YC2022-S692）
Natural Science Foundation of Jiangxi Province(20224BAB202036) （20224BAB202036）
Science and Technology Key Project of Education Department of Jiangxi Province(GJJ2200805) （GJJ2200805）
Innova-tion Fund Designated for Graduate Students of Jiangxi Province(YC2022-S692) （YC2022-S692）
