智能科学与技术学报 ›› 2023, Vol. 5 ›› Issue (1): 92-103.doi: 10.11959/j.issn.2096-6652.202303

• 学术论文 • 上一篇    下一篇

Rapider-YOLOX:高效的轻量级目标检测网络

顾宙瑜, 於跃成, 者甜甜   

  1. 江苏科技大学计算机学院,江苏 镇江 212000
  • 修回日期:2022-11-17 出版日期:2023-03-15 发布日期:2023-03-01
  • 作者简介:顾宙瑜(1996- ),男,江苏科技大学计算机学院硕士生,主要研究方向为计算机视觉、深度学习
    於跃成(1971- ),男,博士,江苏科技大学计算机学院副教授,主要研究方向为计算机视觉、深度学习
    者甜甜(1997- ),女,江苏科技大学计算机学院硕士生,主要研究方向为计算机视觉、深度学习
  • 基金资助:
    国家重点研发计划资助项目(2018YFC0309104);江苏省建设系统科技项目(2021JH03);2022年度江苏省工业和信息产业转型升级专项资金-关键核心技术(装备)攻关产业化项目:“深远海智能风电运维母船研制关键技术攻关”(CMHI-2022-RDG-004)

Rapider-YOLOX: lightweight object detection network with high precision

Zhouyu GU, Yuecheng YU, Tiantian Zhe   

  1. College of Computer Science, Jiangsu University of Science and Technology, Zhenjiang 212000, China
  • Revised:2022-11-17 Online:2023-03-15 Published:2023-03-01
  • Supported by:
    The National Key Research and Development Program of China(2018YFC0309104);The Construction System Science and Technology Project of Jiangsu Province(2021JH03);The Construction System Science and Technology Project of Jiangsu Province: "Key Technology Research for Development of Intelligent Wind Power Operation and Maintenance Mothership in Deep Sea"(CMHI-2022-RDG-004)

摘要:

作为一种轻量级网络结构,YOLOX-Nano具有运行速度快的优势。然而,该网络在实际应用中仍然存在特征提取能力较弱、检测精度不足的缺陷。为此,提出了一种综合平衡检测速度和检测精度的高效目标检测网络Rapider-YOLOX。首先,设计高效瓶颈模块,以提升原始 YOLOX-Nano 模型中深度卷积模块的特征提取能力。其次,设计软空间金字塔池化模块,以避免原始SPP模块容易出现丢失部分重要信息的现象,进一步提升多尺度信息融合及通道间信息交流的能力。最后,引入CIoU损失,利用预测框与真实框的中心距离及宽高比提升预测框的位置精度。在PASCAL VOC2007数据集上的实验结果表明,提出的Rapider-YOLOX模型的mAP达到77.92%,比原始YOLOX-Nano高3.79%。此外,在CUDA核心数仅为384的GT1030上,FPS达到45.40,在CPU上FPS也可达到23.94,从而在确保网络轻量级特性的同时,进一步提升了网络的检测精度和泛化性能。

关键词: 目标检测, 高效卷积神经网络, YOLOX-Nano, 轻量级, 高精度

Abstract:

As a lightweight network structure, YOLOX-Nano has the advantage of fast running speed.However, the model still has the defects of weak feature extraction ability and insufficient detection accuracy in practical application.Therefore, an efficient object detection network Rapider-YOLOX which comprehensively balanced the detection speed and detection accuracy was proposed.Firstly, the highly efficient bottleneck module was designed to improve the feature extraction capability of depthwise convolutional blocks in the original YOLOX-Nano model.Secondly, the soft-SPP module was designed to avoid the loss of some important information in the original SPP module and improve the ability of multi-scale information fusion and information exchange between channels further.Finally, CIoU was introduced to improve the position accuracy of the prediction box by using the center distance and aspect ratio between the prediction box and the real box.The experimental results on PASCAL VOC2007 dataset showed that the mAP of Rapider-YOLOX model reached 77.92%, which was 3.79% higher than the original YOLOX-Nano.In addition, on GT1030 with only 384 CUDA cores, the FPS of the proposed method could reach 45.40.The FPS could also reach 23.94 on the CPU, which further improved detection accuracy and generalization performance of the network while ensuring the lightweight characteristics of the network.

Key words: object detection, efficient convolutional neural network, YOLOX-Nano, lightweight, high precision

中图分类号: 

No Suggested Reading articles found!