基于改进特征金字塔的目标检测方法
Object Detection Method Based on Improved Feature Pyramid Networks
张天飞 1周荣强 2龙海燕 1丁娇 1张磊1
作者信息
- 1. 安徽信息工程学院 电气与电子工程学院,安徽 芜湖 241000
- 2. 杭州智棱科技有限公司,浙江 杭州 310000
- 折叠
摘要
为了缓解多尺度目标特征信息不足的问题,受BiFPN(Bi-directional Feature Pyramid Network)网络的启发,在网络模型的Neck部分提出了一种反转N型特征金字塔结构即IN-FPN(Invert N-Feature Pyramid Network),对网络的多尺度特征融合结构加以优化,其带有侧向连接的层次结构,将特征经过 2 次自上而下和 1次自下而上的双向融合,使得物体的浅层和深层特征充分融合,相互促进.同时,考虑不同尺度特征融合时的贡献不同,给每一个尺度添加可自适应学习权重ωi.此外,为了解决网络退化问题,进一步提升网络性能,参考残差网络结构,增加了含有Block模块的路径.实验结果表明,所提方法在COCO 2017 数据集和VisDrone 2019 数据集上其平均精度(AP)值分别达到了53.02%和25.21%,比基准模型均有所提升,验证了该方法的有效性.
Abstract
To solve the problem of insufficient feature information for multi-scale targets,inspired by the BiFPN network,we proposed IN-FPN,an inverted N-type FPN structure in the Neck section of the network mod-el.In addition,the multi-scale feature fusion structure of the network was optimized,with a horizontally connect-ed hierarchical structure.The features are fused twice from top to bottom and once from bottom to top in a bidi-rectional manner,allowing shallow and deep features of the object to fully fuse and promote each other.At the same time,for the different functions of feature fusion at different scales,adaptive learning weights are added to each scale ωi.Moreover,in order to solve the problem of network degradation and improve network performance,a path containing a Block module was added based on the residual network structure.The experimental results showed:With this method,the AP values reached 53.02%and 25.21%on the COCO 2017 dataset and Vis-Drone 2019 dataset.They were both improved,compared with the benchmark model,verifying the effectiveness of the method.
关键词
目标检测/特征金字塔/多尺度融合/检测精度Key words
object detection/FPN/multi scale fusion/detection accuracy引用本文复制引用
基金项目
安徽省高校自然科学研究重点项目(2023AH052917)
芜湖市科技计划重点研发项目(2022yf64)
出版年
2024
NETL
NSTL
维普
万方数据