多任务特征融合的CenterNet运动车辆检测方法  

Multi-task feature fusion for moving vehicle detection based on CenterNet

作　　者：李晓晗 刘石坚 邹峥 戴宇晨 LI Xiao-han;LIU Shi-jian;ZOU Zheng;DAI Yu-chen(College of Computer Science and Mathematics,Fujian Provincial Key Laboratory of Big Data Mining and Applications,Fujian University of Technology,Fuzhou 350118,China;College of Computer and Cyber Security,Fujian Normal University,Fuzhou 350117,China)

机构地区：[1]福建理工大学计算机科学与数学学院福建省大数据挖掘与应用技术重点实验室,福建福州350118 [2]福建师范大学计算机与网络空间安全学院,福建福州350117

出　　处：《陕西科技大学学报》2024年第5期206-213,224,共9页Journal of Shaanxi University of Science & Technology

基　　金：国家自然科学基金项目(62172095);福建省科技厅自然科学基金项目(2022J01932);福建省教育厅科技计划项目(JAT210283,JAT220052);福建省创新资金项目(2022C0022)。

摘　　要：基于深度学习技术的运动车辆检测是交通和计算机学科当下的研究热点.针对动态车辆检测任务中多尺度、目标重叠、难以区分动态和静态的车辆等难题,本文提出了一种多任务特征融合的CenterNet运动车辆检测方法.首先向网络中新增一支用于实现车辆分割的任务流,与原有目标检测流共同组成双流机制,然后使用恰当的方式实现双流特征融合,辅助增强目标检测流中的关键特征信息,此外,注意力机制的加入进一步优化了模型精度.在以公共数据集UA-DETRAC为基础所制作的测试集上,本文方法的平均精确率为70%,相比原始CenterNet模型提高了5.8%;帧率为30 f/s,在对比方法中具有最佳的速度与精度均衡性.大量实验表明,本文方法能够较好地胜任运动车辆的检测任务.Motion vehicle detection based on deep learning technology is currently a research hotspot in the intersection of traffic and computer science.To address challenges in dynamic vehicle detection tasks,such as multi-scale issues,overlapping targets,and the difficulty of distinguishing between dynamic and static vehicles,this paper proposes a multi-task feature fusion approach for CenterNet motion vehicle detection.Firstly,a task branch for vehicle segmentation is added to the network,forming a dual-stream mechanism along with the original object detection stream.Subsequently,an appropriate method is employed to achieve feature fusion between the two streams,assisting in enhancing critical feature information in the object detection stream.Additionally,the introduction of attention mechanisms further optimizes model accuracy.On a test set created based on the UA-DETRAC public dataset,our proposed method achieves an average precision of 70%,representing a 5.8%improvement compared to the original CenterNet model.With a frame rate of 30 frames per second,our method demonstrates the best balance between speed and accuracy compared to the contrastive methods.Extensive experiments indicate that our approach performs well in motion vehicle detection tasks.

关 键 词：运动车辆检测 分割 CenterNet 多任务学习 特征融合 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]