基于改进RT-DETR的夜间低光照行人检测算法
打开文本图片集
关键词：行人检测;RT-DETR；低光照；注意力机制；多尺度中图分类号：TP391.4 文献标识码：A doi：10.37188/CJLCD.05-005 CSTR：317.14.CJLCD.05-005
Improved RT-DETR-based pedestrian detection algorithm for low-light conditions at night
LU Yan 1，2 ，LIFu 2，3∗ ， QI Mingrui²， YANG Xinmeng²
（1.School of Computer，Nanjing University ofInformation Science and Technology， Nanjing 210044，China;
2.School of Internet of Things Engineering，Wuxi University，Wuxi 214lO5，China;
3. Key Laboratory of Embedded System and Seruice Computing Ministry of Education， Tongji University， Shanghai 2Ol8O4，China）
Abstract：Pedestrian detection in low-light night-time environments faces challenges including high false positive rates，significant false negatives，and insuficient recognition accuracy. To address this，this paper proposes a detection algorithm based on an improved RT-DETR，achieving precise detection under low illumination through multi-module collaborative design. The algorithm embeds an FDT module at the top layer of the feature pyramid，employing a two-stage atention mechanism to enhance weak feature extraction and global context modeling capabilities. A DySample module is deployed in the neck network，employing a dynamicaly learnable spatial resampling mechanism to achieve multi-scale feature alignment and small object detection enhancement. Furthermore，the DRBC3 module serves as the feature extraction core，integrating multi-expansion-rate convolutions and re-parameterisation techniques to construct multi-scale receptive fields，thereby enhancing the capture of details in blurred and occluded objects. Experiments on the LLVIP dataset demonstrate that this algorithm achieves a 1.39% increase in mAP0.5，a 2.21% rise in Precision， and a 3% improvement in Recall，while simultaneously reducing the number of parameters. Inference speed is also significantly enhanced.Generalisation experiments on the NightSurveillnce and Nightowls datasets further validate its superior performance.The algorithm effctively improves detection accuracyand reduces false negatives while maintaining real-time capability，exhibiting robust and practical performance.
Key words： pedestrian detection;RT-DETR； low light;attentional mechanism;multi-scale
1引言
行人检测作为计算机视觉领域的核心任务，旨在从图像或视频中快速识别并精确定位行人，在安防监控、工业巡检等场景中应用广泛。（剩余17441字）
