面向铁路客运场景的对抗鲁棒性人头检测模型  

Adversarial robust head detection model oriented to railway passenger transport scenes

作　　者：随玉腾 戴琳琳 朱宇豪 景辉 SUI Yuteng;DAI Linlin;ZHU Yuhao;JING Hui(Beijing Jingwei Information Technologies Co.Ltd.,Beijing 100081,China;Institute of Computing Technologies,China Academy of Railway Sciences Corporation Limited,Beijing 100081,China;Postgraduate Department,China Academy of Railway Sciences,Beijing 100081,China)

机构地区：[1]北京经纬信息技术有限公司,北京100081 [2]中国铁道科学研究院集团有限公司电子计算技术研究所,北京100081 [3]中国铁道科学研究院研究生部,北京100081

出　　处：《铁路计算机应用》2023年第6期14-19,共6页Railway Computer Application

基　　金：北京经纬信息技术有限公司科研项目基金(DZYF21-36)。

摘　　要：基于人头检测的人群数量估计算法能为铁路客运车站应对突发客流、防止人群聚集提供有效的决策辅助,但人头检测使用的深度学习模型易受到对抗样本影响。为提升深度学习模型的对抗鲁棒性,建立了基于RetinaNet算法的人头检测模型;在Brianwash数据集上分别使用快速梯度符号法(FGSM,Fast Gradient Sign Method)和投影梯度下降(PGD,Projected Gradient Descent)2种对抗攻击方法生成对抗样本,初始模型在对抗样本数据集上的mAP值均有显著下降,验证了对抗攻击对模型性能影响的有效性;再对模型进行对抗训练,对抗训练后的模型在各类对抗样本验证数据集上的mAP值均有显著提升。实验结果表明,对抗训练后的人头检测模型能有效防御对抗样本的攻击,提升模型检测性能和对抗鲁棒性。The crowd estimation algorithm based on head detection can provide effective decision-making assistance for railway passenger stations to cope with sudden passenger flow and prevent crowd aggregation,but the deep learning model used for head detection is easily affected by adversarial samples.To improve the adversarial robustness of deep learning models,this paper established a head detection model based on the RetinaNet algorithm.The paper used two adversarial attack methods,Fast Gradient Sign Method(FGSM)and Projected Gradient Descent(PGD),to generate adversarial samples on the Brianwash dataset.The initial model had a significant decrease in mAP on the adversarial sample dataset,was verified the effectiveness of adversarial attacks on model performance.After conducting adversarial training on the model,the mAP values of the trained model were significantly improved on various adversarial sample validation datasets.The experimental results show that the head detection model trained in adversarial training can effectively defend against attacks from adversarial samples,improve the model's detection performance and adversarial robustness.

关 键 词：目标检测 对抗攻击 人头检测 图像处理 卷积神经网络(CNN) 

分 类 号：U293.13[交通运输工程—交通运输规划与管理]