电子学报 ›› 2023, Vol. 51 ›› Issue (2): 355-364.DOI: 10.12263/DZXB.20210559

• 学术论文 • 上一篇    下一篇

基于特征膨胀卷积模块的轻量化技术研究

许新征1,2, 李杉1   

  1. 1.中国矿业大学计算机科学与技术学院,江苏 徐州 221116
    2.教育部矿山数字化工程中心,江苏 徐州 221116
  • 收稿日期:2021-05-05 修回日期:2021-07-21 出版日期:2023-02-25
    • 作者简介:
    • 许新征 男, 1980年8月生, 安徽宿州人,博士, 教授.主要从事机器学习与数据挖掘、 人工智能与模式识别、 医学图像处理等方面的研究.E-mail: xxzheng@cumt.edu.cn
      李 杉 男, 1995年8月生, 湖北松滋人, 硕士研究生, 主要从事深度学习和计算机视觉等方面的研究.
    • 基金资助:
    • 国家自然科学基金(61976217)

Research of Lightweight Convolution Neural Network Based on Feature Expansion Convolution

XU Xin-zheng1,2, LI Shan1   

  1. 1.School of Computer Science and Technology, China University of Mining and Technology, Xuzhou, Jiangsu 221116, China
    2.Engineering Research Center of Mining Digital Ministry of Education, Xuzhou, Jiangsu 221116, China
  • Received:2021-05-05 Revised:2021-07-21 Online:2023-02-25 Published:2023-04-14
    • Supported by:
    • National Natural Science Foundation of China(61976217)

摘要:

本文从卷积神经网络模型的网络结构入手,利用特征复用的思想,设计了高效的特征膨胀卷积模块.该模块减少了标准卷积模块的输出通道数,引入了多分支结构.通过各个分支上的廉价操作对标准卷积操作的输出特征图进行变换和融合,产生新的特征图.模块的最终输出由各个分支上生成的特征图进行合并连接得到.特征膨胀卷积模块利用特征复用思想复用模型中的特征,在降低模型计算量的同时,丰富了特征图隐含的信息,提高了模型的性能.最后,将特征膨胀卷积模块代替标准卷积模块,设计了轻量化的VGG16(Visual Geometry Group 16-Layer)模型和残差结构,并且在CIFAR数据集和ILSVRC2012(ImageNet Large Scale Visual Recognition Challenge 2012)数据集上取得了较好的分类效果.

关键词: 卷积神经网络, 轻量化, 特征复用, 特征膨胀卷积, 深度学习, 图像分类

Abstract:

This paper starts with the network structure of convolution neural network model, and uses the idea of feature reuse to design an efficient feature expansion convolution module. The module reduces the number of output channels of standard convolution module and introduces multi branch structure. Through the cheap operation on each branch, the output feature map of standard convolution operation is transformed and fused to generate a new feature map. The final output of the module is obtained by merging the feature graphs generated on each branch. The feature expansion convolution module uses the idea of feature reuse to reuse the features in the model, which not only reduces the calculation of the model, but also enriches the hidden information of the feature graph and improves the performance of the model. Finally, the feature expansion convolution module is used to replace the standard convolution module, and the lightweight VGG16 (Visual Geometry Group 16-Layer) model and residual structure are designed, and good classification results are achieved on CIFAR and ILSVRC2012 (ImageNet Large Scale Visual Recognition Challenge 2012) datasets.

Key words: convolutional neural network, lightweight, feature reuse, feature expansion convolution, deep learning, image classification

中图分类号: