摘要
在原CenterNet算法中,以Hourglass为Backbone的目标检测模型平均精度均值高于one-stage算法,但检测速度较低。为此,基于原有CenterNet目标检测算法,对Hourglass-104模型进行改进,设计一种Hourglass-208模型,并给出双特征金字塔网络特征图融合方法。在此基础上对目标大小和训练采用smoothL1损失函数,提出一种新的可端到端训练的目标检测算法T_CenterNet。在MSCOCO数据集上的实验结果表明,该算法目标检测的评估指标AP_(50)、AP_(S)、AP_(M)分别为63.6%、31.6%、45.8%,检测速度达到36frame/s,综合性能优于原CenterNet算法。
In the original CenterNet algorithms,the object detection model with Hourglass as Backbone has a higher mean Average Precision(mAP)than other one-stage algorithms,but it is limited by the low detection speed.To address the problem,a new model named Hourglass-208 is proposed by using the original CenterNet object detection algorithm to improve the Hourglass-104 model.Additionally,a feature map fusion method for Twin Feature Pyramid Networks(TFPN)is given.On this basis,smooth L1 is used for the loss function of the object size to establish a new object detection algorithm,T_CenterNet,which can perform end-to-end training.Experimental results on the MS COCO data set show that the target detection evaluation index AP_(50),AP_(S),AP_(M) of the proposed algorithm are 63.6%,31.6%,45.8%,respectively,and the detection speed of the algorithm reaches 36 frame/s.The comprehensive performance of the proposed algorithm is better than that of the original CenterNet algorithm.
作者
石先让
苏洋
提艳
宋廷伦
戴振泳
SHI Xianrang;SU Yang;TI Yan;SONG Tinglun;DAI Zhenyong(College of Energy and Power Engineering,Nanjing University of Aeronautics and Astronautics,Nanjing 210001,China;Chery Advanced Engineering&Technology Center,Wuhu,Anhui 241006,China)
出处
《计算机工程》
CAS
CSCD
北大核心
2021年第9期240-251,共12页
Computer Engineering
基金
安徽省发改委重大研发项目“面向智能网联汽车的全线控底盘开发及测试验证”。
