陕西科技大学学报2024，Vol.42Issue(5)：206-213,224,9.
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
摘要
Abstract
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 seg-mentation 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 ob-ject detection stream.Additionally,the introduction of attention mechanisms further optimi-zes model accuracy.On a test set created based on the UA-DETRAC public dataset,our pro-posed method achieves an average precision of 70％,representing a 5.8％improvement com-pared 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 meth-ods.Extensive experiments indicate that our approach performs well in motion vehicle detec-tion tasks.关键词
运动车辆检测/分割/CenterNet/多任务学习/特征融合Key words
moving vehicle detection/segmentation/CenterNet/multi-task learning/feature fusion分类
计算机与自动化引用本文复制引用
李晓晗,刘石坚,邹峥,戴宇晨..多任务特征融合的CenterNet运动车辆检测方法[J].陕西科技大学学报,2024,42(5):206-213,224,9.基金项目
国家自然科学基金项目(62172095) （62172095）
福建省科技厅自然科学基金项目(2022J01932) （2022J01932）
福建省教育厅科技计划项目(JAT210283,JAT220052) （JAT210283,JAT220052）
福建省创新资金项目(2022C0022) （2022C0022）
