Long-tail Object Detection Method in Transportation Scene Images Based on improved RT-DETR

Authors

• Mingjun Deng, School of Transportation Engineering, East China Jiaotong University, Nanchang 330013, China; Jiangxi Provincial Key Laboratory of Comprehensive Stereoscopic Traffic Information Perception and Fusion, East China Jiaotong University, Nanchang 330013, China
• Peihua Nie, School of Transportation Engineering, East China Jiaotong University, Nanchang 330013, China
• Junhua Guo, Jiangxi Provincial Key Laboratory of Comprehensive Stereoscopic Traffic Information Perception and Fusion, East China Jiaotong University, Nanchang 330013, China; Jiangxi Flight University, Nanchang 330088, China
• Ningbo Gao, Jiangsu Zhicheng Huining Transportation Technology Co., Ltd., Nanjing 210014, China

Abstract

Abstract: Detecting rare objects in transportation environments is crucial for the safety and reliability of autonomous driving systems. To address the challenges of limited detection capability on rare objects that appear only a few times during training, we propose an improved long-tailed object detection method for traffic scene images named TF-DETR (tail free DETR) based on RT-DETR model. First, the lightweight CSP-EDA (CSP-efficient depthwise aggregation) was constructed to reconstruct the backbone network. Its core unit, the efficient depthwise aggregation, integrated the pooled-transpose attention and convolutional gated linear unit to enhance feature representations for tail-class samples. Second, the spatially-adaptive dynamic cognitive interaction was deployed in the encoder. Through feature reuse between the two-stage, the interaction module established an efficient information interaction mechanism. The mechanism spanned from feature extraction to multi-scale cognitive enhancement for tail-class objects. Finally, the pinwheel-shaped convolution was introduced for feature map downsampling. Through its collaborative design of asymmetric padding and multi-directional kernels, the model enhanced detection capability for tail-class objects while maintaining a lightweight architecture. Experimental results on the CODA2022 dataset show that TF-DETR achieves a 4.9% increase in AP50 and a 5.4% increase in AR50 compared with the baseline. Meanwhile, a 22.9% reduction in FLOPs and a 25.7% decrease in Parameters are realized.

Recommended Citation

Deng, Mingjun; Nie, Peihua; Guo, Junhua; and Gao, Ningbo (2026) "Long-tail Object Detection Method in Transportation Scene Images Based on improved RT-DETR," Journal of System Simulation: Vol. 38: Iss. 9, Article 6.
DOI: 10.16182/j.issn1004731x.joss.26-0200
Available at: https://dc-china-simulation.researchcommons.org/journal/vol38/iss9/6

First Page

2518

Last Page

2534

CLC

U495; TP183

Recommended Citation

Deng Mingjun, Nie Peihua, Guo Junhua, et al. Long-tail Object Detection Method in Transportation Scene Images Based on improved RT-DETR[J]. Journal of System Simulation, 2026, 38(9): 2518-2534.

DOI

10.16182/j.issn1004731x.joss.26-0200